You are on page 1of 889
CIRM TCIM TaL STS STEEL STRUCTURES DESIGN AND BEHAVIOR re ee ecru arate wegen Charles G. Salmon John E. Johnson ee wa MeV eV Ety Table of Most Commonly Used Symbols a = fier wet dimension; ie. weld size. in. (Chap 5; eng (mension pares oad) of pate (Fig. 6147): VEG (Gd = 1/4¢Chap. Ssillener spacing (Chap. 11: deh of ectangular compression sess dstbtion -omerete la Fig 167.1) o AulAp. fai of woh erosional area Ay (0 cross-section aca Ay of compression Oange (see sec. 115) a ~ area oF the cros-sootion: area Abe = croxcsctona area of hase meal ay = gross ers seston wea of bo three rod area hosed on major diameter Ae effective net ne A of fension member ay = ons are of one ange: A ~ compression Mange area (See. 114) 4 "= gras are of ros section Ay sro atca subject shear yield See. 3.8) a fet ates, “tensile sess area” through thveaded portion of tension ced: net area though holes on tension member Aw nt ar ated upon by sear (ce Sox. 3.8) a = nt tea ated upon by tension Se See. 26) ae sich ars fr cll shaped beams: ly For patie 6 “wid: beam ith Mange width dinesion of pate perpendicular o lod tection Fig. 6.14.7) eiective width of ustitfcned compression element (Se. 6 18)flectve la width (ap. 16) Pedy fhange whith, pally for stot W section: for Ream yy babe wiih of sttfener 6 (8/9 yy. cole for practical design: Ey. 1212.8 ® {8/9} fy coco for practical design; Eg. 1212.3 hy = haanifeatonfstr for member in raved frame (no lateral teansltion at ends of member). Eq, 12.103 ‘nication fat for way analy of member io unbraced rae. Exs12.11.9 1251.11 isunce from noua) axis to enone fiber where sWess 56 computed; measured in «- oF wet ‘secording 0 subseip CucCunsCn = fate 1 account for morn pact io eam seg, Egy 9611 ant 9612 with rapt Peds ‘hoot oaks, oro t econ subs cr in moment magniicain relating to moment gradient ani end esr (fined in Sees. 1210 ae nD an 124 Tor ASD, aso Tale 12.3.1, with espoet to beading about x yates. geen e500 subscript Co CnusCns vg = terty = wb shear cosfihent ‘= torsional warping constant (Chup. 8; compression force in web of wee! section (Chap. 16) ‘= overall depth of section: nomial bok diameter 1 Ginance from outside of ange to PNA (Chap. 16) = EP /{12( — x2)] = flexural righty per unit length of plate (Chap. 6 Part Ml and fg, 11.11.14 + ‘width of fntener hoe tobe deducted (Se. 34; numberof 1/16s ofan inch in weld size ‘= factor to account for eccentric load on tiflenes, gs 1.9.28 and 11.93.38 = exceoticity of fod = tension—compression modulus of elasticity = modulus of elasticity of concrete = (w')V7z where w is the density of concrete in pef (ie. 148 ef for normal-weight concrete and fis in ksi, ‘= modulus of elasticity of steel, 29,000 ks = effecve stiffness of «composite section = sees due to direct shear, PA = saess due to torsional moment, 77/1 {= reuired axial stress a point sing LRFD or ASD load combination ‘= required Bening stress ta point using LRED or ASD load combination ‘= compresiontension ses de to estan of warping (aera bending of flanges), (Ebh /4)(d°6 ad?) ‘= specified 28-day compeesive strength of concrete = required tensile tes ta point using LRED or ASD load combination = factored normal (compression or tension) sires ‘equed shear sress at point using LRFD er ASD load combination = allowable Gexural stress jominalsrength ofthe base metal per unit area = critical stress in compression: buckling sess = elastic ercl buckling stress = Bq 121411 = weld metal tensile strength = bolt nominal tensile srength ‘= bolt nominal shear ength residual tess © O3F, = factor of safety ‘= teal strength of structural steel; tensile srength of base metal (Chap. 5) = tenile trengh of bot material = design stvse limit on R/As in tension on bolt subject to combined shear and tension, values in Table 414.1 = yield sees; for beam Mange, Fy: for column web, Fj: for Mange, Fy: for longitudinal reinforcing ‘ar, Fp for siffeners Fry for tension Mange, Fy for web or weld metal, Fy, = gage distance fr fastener holes measured tansvese to direction of load ‘= shear modolus of elasticity, £/(2(1 + p)] ‘unsuppored web height (For unsymmetrical sections it relates t0 the compression side of the neutral ani) (See AISC B42} overall depth (Fig. 6.13.1): distance between centers of anges (Chap, 8): depth of wed plate Chap 11): story height (Chaps. 14 and 15) twice the distance ftom the neutral axis to the inside face of the compression flange less the fillet or comer radius = eixance between Mange centzoigs ee factor (Eq 49.1) = factored lateral force casing sway defection ‘moment of inert bout the x of as, respectively ‘= moment of nena of one flange about y-axis (Chap. 8) * girder moment of inna (Chaps. 14 and 15) ‘= polar momest of neta about shear center wow Mi Ma. Me = maient of ie be rss sectional ae of taniene see 3 eine flowing, 1.1141 Trformed cracked section momen of inet of composite sein (Chap 16) = prettier. fn Sec. 7.10) Iroment of neni of compression ange so 8 LET secs 62 and 122, dsance from outer face of lange to web toe of Filet (Sx. 18: par SNeking ceticim bg 51428: png conan Sec. 9.13 spring consam = ‘Shh 139 = cei nis er ald spe 4g Tables 74.1 a 962) = ete ent ator Se 69 mh spel othe» oF aes respectively = fraliied ete lng actor wed in desgn of colin supporting a ane” colon, E1411 ‘Hecne length free Jo plane of bending with o Ine UaRsltion and wid sdeway, rape ‘hse 12 = sseraipuntvaced erath ar sain the icon of the force (Chap. 4) 0 ditancs easurad in dren of tne offrce (Chap. 4 * AISC! asirum erally uibrced length for using My = Mp. Eq 963 ally unbraced eng for using plastic analysis, F4 9.6.2 AIC! maximum terally uneaced length for using My = M, = O.7F,S, length of File aehs AAISC! ax a ‘numberof shar planes ‘Chap 4) uniformly distributed service load torsional moment (Chap. 8) = faceted uiformls disburse torsional moment (Chap. 8) smaller toment Mand lager moment Ms athe ends Of aterally unbraced segment ‘= requzed moment using ASD Jd combination ‘moment ‘absolute saluey at 1/4, 1/2. and V8 points. respectively, of laterally unbraced some. Eq 9.5.1 taste later corsionl ching moment strength fsuialent constant moment. CyuM for beamcolumn in braced frame subjert 10 cad momen Mi and Mz only (See. 12.4, Jarl bending moment on ne Mange, (EC /h) dd) Primary bending moment from first-der analysis firs-onder moment in sway analysis under H, fetce ‘maximum moment inthe unbraced segment (Ser 9.6) nominal moment strength ominl moment strength nthe presence of shear Primary facored moment forthe no translation case of beam-> le gathered on such a roof, Furthermore, though wind is frequently ignored asa vertical force on a roof, nevertheless it may cause such an effect, For these reasons, a 20 psf (960 2) minimum loading, eventhough it may not always be snow, i reasonable. Local col actual weather conditions, ASCE 7, or the Canadian Structural Commentaries Port 4 1.6}, should be used when designing for snow. Other snow load information has been provided by Lew, Simi, and Flingw00d the Ruilding Structural Design Handbook (Chapter 2) 17), and inthe works of O'ROUKE and Stiefel (1.36), Temptin and Schriever {1.37}, O"Rouske, Tobasson, and Wood (1.35 (O'Rourke, Redfield, and von Bradsky 139], O'Rourke, Speck, and Stet! [10)- Sack 11.41], O'Rourke and Galanais (1.42) and Sack and Gives {1 ask Section 14 Loads 27 Wind Load {All structures are subject to wind load, but usualy only those more than three or four sto- ries high, as well as long bridges, require special consideration. ‘On any typical building of rectangular plan and elevation, wind exerts pressure on the windward side and suetion on the leeward sie, as well as either uplift or downward pressure on the roof. For most ordinary situations, vertical roof loading from wind is Deplected on the assumption that snow loading will require a preater strength than wind Toading, Ths assumption is not true for southern climates where the vertical loading due to ‘wind must be included. Furthermore, the total lateral wind load, windward and leeward effect, is commonly assumed to be applied t the windward face ofthe building ‘in accordance with Bernoulli's theorem for an ideal fluid striking an object, the increase in static pressure equals the decrease in dynamic pressure, given by L eI 143) a= 50" 43) where gis the dynamic pressure on the object, pis the mass density of air (specific weight to = 0007651 pet at ea level and 15°C), and Vs the wind velocity In terms of velocity V in miles per hou, the dynamic pressure (psf) would be 1 (007681 /S280V? _ 5 yas y20 . 2 3(5)(m) = 90m" (uaa [ASCE 7 defines the velocity pressure using @ modified form of the above equation: a4sy 4, = 0100256 KK KV" where K, i the wind directionality factor, isthe velocity pressure expos isthe topographic factor, Tis the importance factor, and ‘qi the calculated g, at midhefght ofthe roof. © coefficient, In design, the dynamic pressure q is commonly converted into equivalent static pres- sure p, which may be expressed {1.2, Sec. 6) P= QGCy ~ a(GCp) 46) where q = gg for windward walls evaluated at height z above the ground = gy for leeward walls, side walls, and roofs 4) = 4 for leeward walls side walls, and roofs of enclosed buildings 41" qe for positive intemal pressure evaluation in panially enclosed buildings G = gust effect factor Cp = external pressure cooficient GC, = internal pressure coefficient ‘ly wth winding systems ony the fst em is come bea the second term cancels out as it produces equal end opposite pressures on the windward and leeward walls. Excellent deals of application of wind loading to strutures are available in the ASCE 7 Standard and inthe National Building Code of Canada {1.6} oo = 0652V%, fr gin MPa and V ine aaa 28 © Chapter 1 Introduction we La Force developed by ea au. For all buildings having nonplanar surfaces, plane surfaces inclined to the wind rection, or surfaces having significant openings, special determination of the wind forces should be made using sources like the ASCE 7 Standard or the National Building Code of ‘Canada [1.6]. For more extensive treatment of wind loads, the reader is refered 0 the Report ofthe Task Committee on Wind Forces |1.33], Lew, Simiu, and Ellingwood in the Building Structural Design Handbook |17, Chap. 2}, Mehta (1.34), and Stathopoules, Sury, and Davenport [1.35], Earthquake Load ‘An earthquake consists of horizontal and vertical ground motions, with the vertical mation ‘sually having the much smaller magnitude. Because the horizontal mation ofthe ground causes the most significant effect, its that effect which is often thought of as earthuake load. When the ground under an object (structure) having mass suddenly moves, the ine. tig of the mass tends to resist the movement, as shown in Fig. 1.4. A shear force is devel ‘oped between the ground and the mass. Most building codes having earthquake provisions requice that the designer either (1) use a dynamic analysis of the structure, or (2) for normal generally rectangul, ‘medium-height buildings, use an empirical lateral base shear force CW. The dynamics of earthquake setion on structures is outside the scope of this text, andthe reader i efeed to Chopra [1.46] and Clough and Penzien [1.47 ‘The equivalent lateral base shear force procedure for designing earthquake-resistance has traditionally been used by most building coves to simplify the design process, For many ‘years, a widely used source has been the Structural Engineers Association of Califia (SEAOC) recommendations [1.44], the latest version of which is 1999. ASCE 7 contains an equivalent lateral force procedure for “Buildings designated as regular up to 240 fet" wherein the seismic base shear Vis given as vecw aan where C, = seismic response coefficient, varying from around 0.01 fora low velo related acecleration coefficient (ay, 0.05) on good soil (say, rock) with goon seismic-resiting structural system (say, a moment-esisting fame) for {he maximum 240-foo-high “regular” building, wo around 0.35 for high velocity: related acceleration cosficient (say, 0.20) on poor soil (say, sft cy ‘or silt) with a poor momentressting system (say, unreinforced masonry fot 2 120-foot-high “regular” building, W = total dead load of the building, including partitions, and portions of other loads as defined in ASCE 7 fr \ 1 Tho. + oF _ ee a Section 14 Loads © 29 The seismic response coefficient is given by pee 4s) RIL ‘where Sps = 5% damped design spectral response acceleration for short periods °R = response modification factor relating to the seismic Force-esstng struc- tural system 1 = occupancy importance factor However, the seismic response coefficient need not be greater than on fort =, (149) FRI = fort > (14.90) PRD, ‘where Sj = design spect esponse parameter at a period of 1.05 r fundamental period ofthe structre 7, = long pero transition pti specif in ASCET for diferent regions Also, C shall ot be less than c= 00 for, < 068 (1492) ass; for = 06% 1494 oo Set (14say where S| mapped maximum spectral response acceleration parameter “The fundamental period of the structure, 7; used in Eqs. (14.9) shall not exceed, Tox ~ Cale (14.10) where Cy = coefficient relating t0 5p), varying from 1.410 1.7 as Spy varies from 0.4 t0 1.7 approximate fundamental period determined from Te > Gis aan \where C, and x are parameters related to the type of structure and fy isthe height of the structure above the base [Note thatthe base shear force method is for “regular” buildings. regular buildings are those which contain (1) plan structural irregularities, such xs torsional irregularity, re- entrant comers, diaphragm discontinuity, out-of-plane offsets, and nonparalel systems, andlor (2) vertical structural irregularities, suchas stiffness iregularty (soft story), mass inregulaity, geometric iregularty in-plane discontinuity in vertical lateral force-ressting elements, and discontinuity i lateral strength (weak story). The base shear force method is Also limited to buildings not excveding 250 fin height. ‘After the base shear force Vi determined, the vertical distribution of seismic forces ‘mast be determined, The seismic design story shear must include direct shear as well as, torsion. The building must be designed to resist overturning effects caused by seismic {otces. Also, story drifts, and where required, member forces and moments due to P-delta, effects must be determined. ASCE 7-08 is based on the NEHRP Recommended Provisions for the Development (of Seismic Regulations for New Buildings {1.49}, whichis the definitive source for seismic 30 © Chapter? Introduction ‘design. Various waditional building codes for earthquske-resistant design are compared by Chopra and Cruz {1.48}, Many states have adopted the International Building Code (18C) 1.48) which contains provisions for seismic design generally based on the ASCE 7 Stan- {ard [1.2], Forstel design, AISC has recently (2006) published the Seismic Design Man. tual (1.53), Other information on stee-telated earthquake codes is provided by Popov (1.51) and Marsh 11.52) 1.5 TYPES OF STRUCTURAL STEEL MEMBERS Figue 151 Standard rolled shapes. As discosod in Sec 1.2, the fanction of structure isthe principal factor determining te structral configuration. Using the suctral configuration along, with the design las, individual components are sel o ropely support and transit loads troaghot ti Structure. Steel members are selected fom song the stands rolled shapes adopted by the American Isitte of Steel Consirition (AISC) (ls given by American Socket fo Tesing and Materials [ASTM] A6 Speciation) Ofcourse, welding permits combing plates andor other rolled shapes to obtain any shape the designer may resi. “Typical roll shes, ie dimensions for which are found inthe ATSC Mana [1.5 ae shown in Fig 151, The most commonly used section is the wide ange shape Fa, 15.1) which fomedby hotoingin he ste il The wide-Range apes designed the nominal depth an the weight per fot sch a WIR which sory 18. op (octal depth = 18.59 in secrding to APSC Mara) and weighs 97 pounds per fox. (at nits the W187 section cull be designated W460%144, meaning nominally 460 rm dep an having amas of 184 Kal. Two sets of dimensions ae found in the ASC Mra nee sated in doima forthe designer ose in computations, and ane set exesedin focsions (the smallest increment) othe deter ose on pans nd shop anes Rolled W shapes ares dexignned by ASTM AG/AGM™ [15] in aoondace with web ik nessasGroup | trough 5, withthe thinnest we sections in Group “The American Staal heim (ig. 15.1), commonly ele the E-bay has sively narrow and sloping flanges anda hick web compared to the wide-Mange sap. Ue ITCet - ‘eam Sent - O =a — “Tite ofa wthe I vesen of be ASTI Stade. Figure 152 ‘Some cold-frmed shapes, of beams has become uncommon because of excessive material inthe web and relative Tack of lateral stfness resulting from the narrow flanges. ‘The channel (Fig. 1 5.1e) and angle Fig. 15.14) are commonly used either alone ot incombination with otic: sections. Te channel s designated, for example, #8 C1220, 4 ominel 12in deep channel having a weight of 20.7 pounds per foot Angles are desig- hated by their lg length (long lg ist) and thickness, such as LOX4%G, “The structural te (Fig, 15. 1e) is made by cutting wide-flange or [beams in half and {is commonly used for chotd members in trusses. The tee is designated, for example, a8 WTS%44, where the 5s the nominal depth and isthe weight in pounds pr foot, this tee being cu from a W108, Pipe sections (Fig. 1.5.10 are designated “standard” “extra strong,” and “double- ‘extra stong” in accordance with the thickness and ae also nominally prescribed by diam eter: thus 10-indiam,dovble-extrastoag is an example ofa particular pipe size. Hollow structural ste! (HSS) (Fig. 1-5.1g) is used where pleasing architectural appearance is desired with exposed sel. Tubing is designated by outside dimensions and thickness, such ss SS 8X63} ‘The sections shown in Fig. 15.1 ae all hotolled: tha is, they are formed from hot billet sicel (blocks of stel) by passing through rolls numerous times to obtain the final shapes. Many other shapes ae coldformed from plate material having a thickness not exceeding Lin, a shown in Fig. 1.5.2 Regarding size and designation of coldformed ste members, there are no truly stan- rd shapes eventhough th properties of many common shapes ae given nthe Cold-Rormed ‘Stel Design Manual 19), Various manufacturers proce many proprietary shapes. (C Lb Lo & SLL Tension Members Tension members commonly occas chord membesin ats x goal racing in types srucars a dest suport for bln, cles in spend oot systems eed 2 sspesion ge nin aes od suspenders ht spot he ready, Type os scons of esion member re own in ig, 153, and hides excep for spel fo to rlaing to supenson ype cable supe stirs) treated n Chapter Compression Members scat comprension ember stength is funtion fhe al hae (di croesctonal sap ans of aration) te ara generally spread ou un muchas praca Chord aber in tases a ary ner colt in ung se examples of memes set 32. © Chapter Introduction O @ J dh (o) Rend an cungle (0) Cates comes (er Saplewnd F Figure 153 (4) Rots We ans (0) Srctst (0 Bains tor Typical tension members, Secon “ = compression. Even under the most ideal condition, pure axial compression is aot atin sb; s, design for “axial” loading assumes the effect of any small simultaneous beading may be neglected. Typical cross-sections of compression members are shown i Fig. 154, and their behavior and design are treated in Chapter 6. fd Shapes ao oo Figure 154 O Typical compression lo tava . ‘members (eset 1) Bait eine Beams ean remembers sobjected to transverse lating and are mest fie when hei is distributed so as to be located at the greatest practical distance from the neutral axis. 7 ‘most common beam sections are the wide-flange (W) and I-beams (S) (Fig. 1.5.58) ‘well as smaller rolled I-shaped sections designated as “miscellaneous shapes” (M): or deeper and thinmr-webed sctions thn an econo be role, #e4) {ahupd secton (15.5) ate ee, ncuding send plac idee Figwe 155 ‘yp beam meres, Section 15 Types of Structural Steol Members. * 33 “WA | a £ ; et Led For moderate spans carrying light loads, open-web “Joists” are often used (Fi. 1.5.5). These are parallel chord truss-type members used forthe support of floors and roofs, The stel may be hot rolled or cld-formed. Such joists are designated “K-Series” table for members having the direct support of floors and roof decks in buildings. The LH-Series and DLH-Series are known as LLongspan and Deep Longspan, respectively. Longspan Steel Joists are shop fabricated tmusses used “forthe direct support of floor or roo! slabs or decks between walls, beams, and main structural merabers” [1.10]. Deep Longspan Joists are used “forthe direct sup ‘ort of roof slabs or decks between wall, beams and main structural members” [1.10]. The design of the chords for K-Seres trusses is based on a yield strength of $0 ksi (345 MP), ‘while the web sections may use ether 36 (248 MPa) or 50 ksi (345 MPa). For the LH-and DL#Series, the chord and web sections design must be based on a yield strength of at least 36 ksi (248 MPa) but not greater than 50 ksi 345 MPa). “The K-Series joists have depths from 8 to 30 ia, for clear spans to 60 ft. The Longspan joists (LH-Series) have depths from 18 to 48 in. for clear spans to 96 ft. The Deep Longspan joists (DLH-Seres) have depths from 52 0 72 in. for clear spans to 144 ft All ofthese joists are designed aocording to Specifications adopted by the Stee Joist Insite (SMD [1.10], which generally agree with the AISC Specificaion {1.13} for hot rolled stels, and with the AIST Specification (1.8 for coldformed steels. Designing with Stee Joists, Joist Girders, Steet Deck by Fihet, West, and Van de Pas [1.11] provides excel- Jen eatment of joists und joist-elated floor systems. Dynamics of structures is outside the scope of hi x and he eer efere to Chop 146] and Chugh and Pein ro For beams (known as lintels) carrying loads across window and door openings, angles are frequently used; and for beams (knowa as girts) in wall panels, channels are frequently used, ater Se 24 fr detinion 34 © Chapter 1 introduction Bending and Axial Load When simultaneous action of tension or compression along with bending occu, com: bined load problem arises and the type of member used will be dependent on the type of load that predominates. A member subjected to axial compression and bending is usually referred toas.a bea-coumn. the behavior and design of which is dealt with in Chapter 12. “The aforementioned illustration of types of members to resist various kinds of lod is mended only o showcommon and representative types of members and not to be all-inhsie 1.6 STEEL STRUCTURES Figure 161 Flor joist plane esses) fn sect docking (Photo by EG Salmon) Structures may be divided into three general categorie: (a) framed structures, where ee rmenls may consist of tension members. columns, beams, and members under combined bending and axial oad: thy shell-type structures. where axial forces predominate; and ‘ct suspension-ype structures. where axial tension predominates the principal suppon system, Framed Structures “Most typical building construction isin this category. The multistory building usually con sists of beams and columns. ether rigidly connected or having simple end connections along with diagonal bracing to provide stability. Even though a multstory building i thyee-dimensional ic usualy is designed to be much stiffer in one ditection than the other: thus it may reasonably be teated asa series of plane frames. However, if the framing is 36 © Chapter Introduction Figure 1.64 ‘Comimueus onhtropic pte glder across Misiupp River at St Louis, Misour (Photo by C.6. Salmon) Fie 163 Conn as dg. uriidge Crossing Sten and, New Yorke ho by E°6: Samo) | Bridges are mosily framed structures such 2s beams an plate girders (se Fig. 164), ‘or trusses, usually continuous (se Fig. 1.6.3}. “This text is devoted to behavior and design of elements in framed structures. ‘Shell-Type Structures, In this typeof structure, the shell serves a use function in adton to participation in a ing loads. One common type where the main stess is tension is the concanment vee sed to store liquids (for both high and low temperatures), of which the elevated wate nk {s a notable example, Storage bins, tanks, and the hulls of ships are other examples. Ot ‘many shelltype structures, a framed structure may be used in conjunction with the shell ‘On walls and flat roofs the “skin” elements may be in compression while they 3 ‘together with a framework, The areraft body is another such example, Shelltype structures are usually designed by a specialist and are not within the sore ofthis tex, Figure 1.66 (ble suspended root for Madison Square Garden Sports tnd Bateraiament Cemer, New York. (Pato ‘ourey Bethleem Stet Cxperation| Figure 1.67 ‘Suspension bide. Golden ‘Gae Bridge, San Francine, Calon: Photo by C. G. Samoa} Section 16 Steal Structures © 37 Suspension-Type Structures In the suspensiontype structure, tension cables are major supporting elements. A roof may ‘be cable-tupported, as shown in Fig. 1.6.6. Probably the most common structure ofthis type isthe suspension bridge, as shown in Fig. 1.6.7. Usually a subsystem of the structure consists ofa framed structure, s in the stiffening truss forthe suspension bridge. Since the teasion element isthe most efficient way of earrying load, stuctures ling this concept are increasingly being used 38° * Chapter 1 Introduction Many unusual structures utilizing various combinations of framed, shelltype, and suspension-type structures have heen buill. However, the typical designer must principally understand the design and behavior of framed structures. 1.7 _ SPECIFICATIONS AND BUILDING CODES Sirdtural tel design of buildings in the United Stats is principally based on the spc cation of the American Institue of Steel Consruction (AISC). AISC is composed of ss fabricator and manufacturing companies, a6 wel a individuals interested in stecl design and research. The AISC Specification is the result of the combined judgment of researchers and practicing engineer. The escarch efforts have been synthesized into practical design ‘procedures 1 provide a safe, economical structure. The advent of the digital computer in design practice has made feasible more elaborate design rules. The curret specication Which is referred to throughout this book isthe 2005 Design Specification for Stractrl Stee! Buildings {1-13}. A specification contining a set of rules is intended to insure safety: howeves the designer must understand the Behavior for which the rule applies. Otherwise, an absurd design may rest ether unsafe or grossly conservative. "The authors contend tha itis vito ally impossible to write rules that fully apply to every situation, Behavioral understanding ‘must come firs; application of rules then follows. No matter what set of rules is applic, the designer has the ultimate responsibilty fora safe structee. ‘A specification when adopted by AISC is actually ast of recommendations pu forth by ahighly respected group of expen inthe field of sel research and desiga. Only when ‘governmental bods, such as vty, sate, and federal agencies, who bave legal respoesibily for public safety, adopt or incorporate a speciation such a8 the AISC Steet Design Spec ‘ation (ANSUAISC 360-05) [1.13] ito thei building codes does it become legally aia ‘The design of see bridges is generally in accordance with specifications of tht American Associaton of State Highway and Trassportaion Officials (AASHTO) [13] This isa legal st of rules because it has been adopted by the states (usally the state bik sway departments have this esponsibility), Railroad bridges ae designed in accordance with the specifications adopted hy tit American Railway Engineering and Maintenance of Way Association (AREMA) [14-18 tis case the raloads have the responsibility for safety and through ther own organization adopt the rales o insure safe designs. ‘The term building codeis sometimes used synonyiously with speciation. Mo conreily a building code is a broadly based document. ether a legal documest eich 352 state or Teal building code, ora document widely recognized eventhough not egal hich covets the same wide range of topics asthe stat or loa! building code. Building coe Ee" erally teat all fcets relating to safety, such as tractral design, aychitectural details, protection, eating and airconditioning, plumbing and sanitation, and lighting-On te OR hand, specifications frequently refer rules set forth by the architect or enginet thet tui lo only one particular bulldng while under constriction, Building eves ls rin) ‘prescribe standard loads for which the structure is to be designed, as discused ia Se. "4 ‘The reader should not be disturbed by the interchangeable use of building cove a4 specification, but should clealy understand that whichis legelly required for dss0 that which could be thought of as recommended practice 1.8 _ PHILOSOPHIES OF DESIGN ‘The 2005 AISC Specification for Structural Steet Buildings provides an integrated sen of Allowable Stength Design (previonly refed to ty ASC a he AOE Section 18 Philosophies of Design * 39 tress Metiod) and Load and Resistance Factor Design. The new specification combines the two design methods and replaces earlier specifications that teated the two design methods separately. Allowable Stress Design has been the principal philosophy used dur ing the past 100 years. During the past 20 years or so strctural design has been moving toward a more rational and probability-based design procedure referred tous “limit states” ‘design, Haajer (1.24) and Kennedy [1.25, 1.26] have presented the current status of the Timit states concept and its use in design, Limit states design includes Uhe methods com» ‘monly refered to as “ultimate strength design,” “strengih design,” “plastic design,” “load fac~ tor design’ "limit design” and the more rent “load and resistance factor design (LRFD).” Structures and structural members must have adequate strength, as well as adequate stiffness and toughness, to permit proper Functioning during the service life of the structure “The design must provide some reserve strength above that which is needed to camry the ser vice loads; that i, the structure must provide forthe possibility of overload. Overloads ean arise from changing the use for which a particular structure was designed, from underesti- ‘mation of the effects of loads by oversimplifcations in structural analysis, and from vari tions in construction procedures. In adition, there must be a provision forthe possibilty of understrength, Deviations in the dimensions of members, even though within acoepred tolerances, ean result in a member having less than its computed strength. The materials (stel for members, bolts, and welds) may have less strength than usd inthe design calew lations, A steel section may occasionally have yield stress below the minimum specified ‘value, but stl within the statistically acceptable Timits, ‘structural design must provide for adequate safety ao matter what philosophy of design is used, Provision must be made for both overload and understrength. The study of ‘what constitutes the proper formulation of structural safety has been continuing during the pest 30 years (1.17, I18]. The main thrust has been to examine by various probabilistic methods the chances of “fsilure” occurring in a member, connection, or system. Rather than “flue,” the term “Limit stato” i prefered. Limit states means “those conditions ofa structure at which it ceases to fulfil its imended function” [1.25]. Limit slates are generally divided into two categories: srength and serviceability. Suength (ie. safety) limit states are such behavioral phenomena as achieving ductile maximum strength Ge. plastic strength), buckling, fatigue, fracture, overturning, and sliding. Serviceability Jimit states are those concerned with occupancy of «building such as deflection, vibration, permanent deformation, and cracking. Both the loads ating on the structure and its resistance (strength) to loads are vari= ables that mast be considered. In general, a thorough analysis of all uncertainties that right influence schieving a “limit state” is not practical, or peshaps even possible. The current approach to a simplified method for obtaining a probability-hased assessment of Steuctual safety uses first-order second:-moment reliability methods [1.29], Sueh methods assume thatthe Toad (or load effet) Q and the resistance R are random variables. Typical ‘requency distributions of these random variables are shown in Fig. 18.1. When the resis tance R exceeds the load (or load effect) Q there will be a margin of safety. Unless ® exceeds Q by a large amount, there will be some probability that R may be less than Q. Structural “failure” (achievement of limit state) may then be examined by compar- ing R with Q. or in logarithmic form observing ln(R/Q), a5 shown in Fig. (8.2. "Failure™ is represented by the cross-hatched region. The distance between the failure line and the ‘mean value of the function {Ta(R/Q)] is defined as a multiple of the standard deviation ‘of the function. The multiplier is called the reliability index. The larger is the greater is the margin of safety ‘As summarized by Pinkham (1.18), the reliability index is useful in several ways: 4, Iran give an indication ofthe consistency of safety for various components and systems using traditional design methods. 2, Itcanbe used to establish new methods which will have consistent margins of set. 40 © Chapter 1 Introduction Figure 18.1 Frequency distiutions of Jad @ and resistance. Figure 18.2 Relbility index 8. Eat 3. Itean be sed to vary ina rational manner the margins of sefety for thos ompo- rents ad systems having a gear or lesser need for safety than that require in rian situations In general the expression for he structural safety requirement may be writen as 48, > [R= 2701] aan vihere the let side of. 1.8.1 represen the resistance, or strength, ofthe component t

You might also like