Toma, S.; Duan, L. and Chen, W.F.

Bridge Structures
Structural Engineering Handbook
Ed. Chen Wai-Fah
Boca Raton: CRC Press LLC, 1999
Bri dge Structures
Shouji Toma
Department of Civil Engineering,
Hokkai-Gakuen University, Sapporo, Japan
Li an Duan
Division of Structures, California
Department of Transportation, Sacramento,
CA
Wai -Fah Chen
School of Civil Engineering,
PurdueUniversity,
West Lafayette, I N
10.1 General
10.2 Steel Bridges
10.3 ConcreteBridges
10.4 ConcreteSubstructures
10.5 Floor System
10.6 Bearings, Expansion Joints, and Railings
10.7 Girder Bridges
10.8 TrussBridges
10.9 Rigid FrameBridges(Rahmen Bridges)
10.10Arch Bridges
10.11Cable-Stayed Bridges
10.12Suspension Bridges
10.13DeningTerms
Acknowledgment
References
Further Reading
Appendix: Design Examples
10.1 General
10.1.1 Introduction
Abridgeisastructurethat crossesover ariver, bay, or other obstruction, permittingthesmooth and
safepassageof vehicles, trains, and pedestrians. An elevation view of a typical bridgeisshown in
Figure10.1. A bridgestructureisdivided into an upper part (thesuperstructure), which consistsof
theslab, theoor system, andthemaintrussor girders, andalower part (thesubstructure), whichare
columns, piers, towers, footings, piles, andabutments. Thesuperstructureprovideshorizontal spans
such asdeck and girdersand carriestrafc loadsdirectly. Thesubstructuresupportsthehorizontal
spans, elevating above the ground surface. In this chapter, main structural features of common
types of steel and concrete bridges are discussed. Two design examples, a two-span continuous,
cast-in-place, prestressed concretebox girder bridgeand athree-span continuous, compositeplate
girder bridge, aregiven in theAppendix.
c 1999by CRCPressLLC
FIGURE10.1: Elevation viewof atypical bridge.
c

1
9
9
9
b
y
C
R
C
P
r
e
s
s
L
L
C
10.1.2 Classication
1. Classication by Materials
Steel bridges: A steel bridgemay useawidevariety of structural steel components
and systems: girders, frames, trusses, arches, and suspension cables.
Concretebridges: Therearetwo primary typesof concretebridges: reinforced and
prestressed.
Timber bridges: Wooden bridgesareused when thespan isrelatively short.
Metal alloybridges: Metal alloyssuch asaluminum alloy and stainlesssteel arealso
used in bridgeconstruction.
2. Classication by Objectives
Highwaybridges: bridgeson highways.
Railwaybridges: bridgeson railroads.
Combinedbridges: bridgescarryingvehiclesand trains.
Pedestrianbridges: bridgescarryingpedestrian trafc.
Aqueduct bridges: bridgessupportingpipeswith channeled waterow.
Bridgescan alternatively beclassied into movable(for shipsto passtheriver) or xed
and permanent or temporary categories.
3. Classication by Structural System(Superstructures)
Plategirder bridges: The main girders consist of a plate assemblage of upper and
lower angesand aweb. H- or I-cross-sectionseffectively resist bendingand shear.
Box girder bridges: The single (or multiple) main girder consists of a box beam
fabricatedfromsteel platesor formedfromconcrete, whichresistsnot onlybending
and shear but also torsion effectively.
T-beambridges: A number of reinforced concreteT-beamsareplaced sideby side
to support theliveload.
Compositegirder bridges:Theconcretedeck slabworksinconjunctionwiththesteel
girdersto support loadsasa united beam. Thesteel girder takesmainly tension,
whiletheconcreteslab takesthecompression component of thebendingmoment.
Grillagegirder bridges: Themain girdersareconnected transversely by oor beams
to formagrid pattern which sharestheloadswith themain girders.
Trussbridges: Trussbar membersaretheoretically considered to beconnected with
pins at their ends to form triangles. Each member resists an axial force, either
in compression or tension. Figure10.1 showsaWarren trussbridgewith vertical
members, which is a trough bridge, i.e., thedeck slab passes through thelower
part of thebridge. Figure10.2 showsacomparison of thefour design alternatives
evaluated for Minato Oh-Hasshi in Osaka, Japan. The truss frame design was
selected.
Archbridges: Thearch isastructurethat resistsload mainly in axial compression.
In ancient timesstonewasthemost common material used to construct magnif-
icent arch bridges. Thereis a widevariety of arch bridges as will bediscussed in
Section 10.10
c 1999by CRCPressLLC
FIGURE10.2: Design comparison for Minato Oh-Hashi, Japan. (FromHanshin Expressway Public
Corporation, ConstructionRecordsof MinatoOh-Hashi, Japan Society of Civil Engineers, Tokyo [ in
Japanese] , 1975. With permission.)
Cable-stayedbridges:Thegirdersaresupportedbyhighlystrengthenedcables(often
composedof tightlyboundsteel strands) whichstemdirectlyfromthetower. These
aremost suited to bridgelongdistances.
Suspensionbridges: Thegirdersaresuspended by hangerstied to themain cables
which hang from thetowers. Theload istransmitted mainly by tension in cable.
c 1999by CRCPressLLC
Thisdesign issuitablefor very longspan bridges.
Table10.1showsthespan lengthsappropriateto each typeof bridge.
4. Classication by Support Condition
Figure10.3showsthreedifferent support conditionsfor girder bridges.
Simplysupportedbridges: Themain girdersor trussesaresupported by amovable
hingeat oneend and axed hingeat theother (simplesupport); thusthey can be
analyzed usingonly theconditionsof equilibrium.
Continuouslysupportedbridges: Girdersor trussesaresupported continuously by
morethan threesupports, resultingin astructurally indeterminatesystem. These
tend to bemoreeconomical sincefewer expansion joints, which havea common
causeof serviceand maintenanceproblems, areneeded. Sinkageat thesupports
must beavoided.
Gerber bridges(cantilever bridge): A continuous bridge is rendered determinate
by placing intermediatehingesbetween thesupports. Minato Oh-Hashisbridge,
shown in Figure10.2a, isan exampleof aGerber trussbridge.
10.1.3 Plan
Beforethestructural design of abridgeisconsidered, abridgeproject will start with planning the
fundamental design conditions. A bridgeplan must consider thefollowingfactors:
1. PassingLineand Location
A bridge, beingacontinuation of aroad, doesbest to followthelineof theroad. A right
anglebridgeiseasytodesign and construct but often forcesthelinetobebent. Askewed
bridgeor acurved bridgeiscommonly required for expresswaysor railroadswherethe
road linemust bekept straight or curved, even at thecost of amoredifcult design (see
Figure10.4).
2. Width
Thewidth of ahighway bridgeisusually dened asthewidth of theroadway plusthat of
thesidewalk, and often thesamedimension asthat of theapproachingroad.
3. Typeof Structureand Span Length
The types of substructures and superstructures are determined by factors such as the
surroundinggeographical features, thesoil foundation, thepassinglineanditswidth, the
length and span of thebridge, aesthetics, therequirement for clearancebelowthebridge,
transportation of theconstruction materialsand erection procedures, construction cost,
period, and so forth.
4. Aesthetics
A bridgeisrequired not only to fulll itsfunction asathoroughfare, but also to useits
structureand formto blend, harmonize, and enhanceitssurroundings.
10.1.4 Design
Thebridgedesign includesselection of a bridgetype, structural analysisand member design, and
preparation of detailed plans and drawings. The size of members that satisfy the requirements
of design codes are chosen [ 1, 17] . They must sustain prescribed loads. Structural analyses are
performed on amodel of thebridgeto ensuresafety aswell asto judgetheeconomy of thedesign.
Thenal design iscommitted to drawingsand given to contractors.
c 1999by CRCPressLLC
TABLE10.1 Typesof Bridgesand ApplicableSpan Lengths
FromJASBC, Manual DesignDataBook, Japan Association of Steel BridgeConstruction, Tokyo (in Japanese), 1981. With permission.
c

1
9
9
9
b
y
C
R
C
P
r
e
s
s
L
L
C
FIGURE10.3: Supportingconditions.
FIGURE10.4: Bridgelines.
10.1.5 Loads
Designersshould consider thefollowingloadsin bridgedesign:
1. Primary loadsexert constantly or continuously on thebridge.
Deadload: weight of thebridge.
Liveload: vehicles, trains, or pedestrians, includingtheeffect of impact. Avehicular
load is classied into three parts by AASHTO [ 1] : the truck axle load, a tandem
load, and auniformly distributed laneload.
Other primary loadsmay begenerated by prestressingforces, thecreep of concrete, the
shrinkage of concrete, soil pressure, water pressure, buoyancy, snow, and centrifugal
actionsor waves.
c 1999by CRCPressLLC
2. Secondary loadsoccur at infrequent intervals.
Windload: atyphoon or hurricane.
Earthquakeload: especially critical in itseffect on thesubstructure.
Other secondaryloadscomeabout with changesin temperature, acceleration, or tempo-
rary loadsduringerection, collision forces, and so forth.
10.1.6 InuenceLines
Sincetheliveloadsbydenition move, theworst casescenarioalongthebridgemust bedetermined.
The maximum live load bending moment and shear envelopes are calculated conveniently using
inuencelines. Theinuencelinegraphicallyillustratesthemaximumforces(bendingmoment and
shear), reactions, and deectionsover asection of girder asaload travelsalongitslength. Inuence
linesfor thebendingmoment and shear forceof asimply supported beamareshown in Figure10.5.
For aconcentrated load, thebendingmoment or shear at section Acan becalculated by multiplying
theload and theinuencelinescalar. For auniformly distributed load, it istheproduct of theload
intensity and thenet areaof thecorrespondinginuencelinediagram.
10.2 Steel Bridges
10.2.1 Introduction
The main part of a steel bridge is made up of steel plates which compose main girders or frames
to support a concrete deck. Gas ame cutting is generally used to cut steel plates to designated
dimensions. Fabrication by welding is conducted in the shop where the bridge components are
prepared beforebeingassembled (usually bolted) on theconstruction site. Several membersfor two
typical steel bridges, plategirder and truss bridges, aregiven in Figure10.6. Thecompositeplate
girder bridgein Figure10.6aisadeck typewhilethetrussbridgein Figure10.6b isathrough-deck
type.
Steel hashigher strength, ductility, and toughnessthan many other structural materialssuch as
concreteor wood, and thusmakesan economical design. However, steel must bepainted to prevent
rustingand also stiffened to prevent alocal bucklingof thin membersand plates.
10.2.2 Welding
Weldingisthemost effectivemeansof connectingsteel plates. Thepropertiesof steel changewhen
heated and thischangeisusuallyfor theworse. Molten steel must beshielded fromtheair toprevent
oxidization. Welding can be categorized by the method of heating and the shielding procedure.
Shielded metal arc welding (SMAW), submerged arc welding (SAW), CO
2
gas metal arc welding
(GMAW), tungsten arc inert gaswelding (TIG), metal arc inert gaswelding (MIG), electric beam
welding, laser beamwelding, and friction weldingarecommon methods.
Therst twoweldingproceduresmentionedabove, SMAWandSAW, areusedextensivelyinbridge
construction dueto their high efciency. Both usean electric arc, which isgenerally considered the
most efcient methodof applyingheat. SMAWisdonebyhandandissuitablefor weldingcomplicated
jointsbut islessefcient than SAW. SAWisgenerallyautomatedandcan beveryeffectivefor welding
simplepartssuch astheconnection between theangeand webof plategirders. Atypical placement
of theseweldingmethodsisshown in Figure10.7. TIGand MIGusean electric arc for heat source
and inert gasfor shielding.
Anelectricbeamweldmust not beexposedtoair, andthereforemust belaidinavacuumchamber.
A laser beam weld can beplaced in air but islessversatilethan other typesof welding. It cannot be
c 1999by CRCPressLLC
FIGURE10.5: Inuencelines.
used on thick platesbut isideal for minuteor artistic work. Sincetheweldingequipment necessary
for heatingand shieldingisnot easy to handleon aconstruction site, all weldsareusually laid in the
fabrication shop.
Theheatingand coolingprocessesduringweldinginduceresidual stressesto theconnected parts.
Thesteel surfacesor partsof thecrosssection at somedistancefrom thehot weld, cool rst. When
thearea closeto theweld then cools, it triesto shrink but isrestrained by themoresolidied and
c 1999by CRCPressLLC
FIGURE10.6: Member namesof steel bridges. (FromTachibana, Y. andNakai, H., BridgeEngineering,
Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
cooler parts. Thus, tensileresidual stresses aretrapped in thevicinity of theweld whiletheouter
partsareput into compression.
Therearetwotypesof weldedjoints: grooveandllet welds(Figure10.8). Thellet weldisplaced
at the junction of two plates, often between a web and ange. It is a relatively simple procedure
with no machiningrequired. Thegrooveweld, also called abutt weld, issuitablefor jointsrequiring
greater strength. Dependingonthethicknessof adjoiningplates, theedgesarebeveledinpreparation
for theweld to allowthemetal to ll thejoint. Variousgrooveweld geometriesfor full penetration
weldingareshown in Figure10.8b.
Inspection of welding is an important task since an imperfect weld may well have catastrophic
consequences. It isdifcult to nd faultssuch asan interior crack or ablowholeby observingonly
thesurfaceof aweld. Manynondestructivetestingproceduresareavailablewhichusevariousdevices,
suchasx-ray, ultrasonicwaves, color paint, or magneticparticles. Theseall havetheir ownadvantages
and disadvantages. For example, thex-ray and theultrasonictestsaresuitablefor interior faultsbut
c 1999by CRCPressLLC
FIGURE 10.7: Welding methods. (From Nagai, N., BridgeEngineering, Kyoritsu Publishing Co.,
Tokyo, Japan [ in Japanese] , 1994. With permission.)
requireexpensiveequipment. Useof color paint or magneticparticles, on theother hand, isacheap
alternativebut only detectssurfaceaws. Thex-ray and ultrasonictestsareused in common bridge
construction, but ultrasonictestingisbecomingincreasingly popular for both its high tech and its
economical features.
10.2.3 Bolting
Boltingdoesnot requiretheskilledworkmanshipneededfor welding, andisthusasimpler alternative.
It is applied to the connections worked on construction site. Some disadvantages, however, are
incurred: (1) spliceplatesareneeded and theforcetransfer isindirect; (2) screwing-in of thebolts
createsnoise; and (3) aesthetically boltsarelessappealing. In special casesthat need to avoid these
disadvantages, theweldingmay beused even for siteconnections.
Therearethreetypesof high-tensilestrength-bolted connections: theslip-critical connection, the
bearing-typeconnection (Figure10.9), and thetensileconnection (Figure10.10). Theslip-critical
(friction) bolt ismost commonlyusedin bridgeconstruction aswell asother steel structuresbecause
it issimpler thanabearing-typebolt andmorereliablethanatensionbolt. Theforceistransferredby
c 1999by CRCPressLLC
FIGURE 10.8: Types of welding joints. (From Tachibana, Y. and Nakai, H., BridgeEngineering,
Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
thefriction generatedbetween thebaseplatesandthespliceplates. Thefriction resistanceisinduced
by theaxial compression forcein thebolts.
Thebearing-typebolt transferstheforceby bearingagainst theplateaswell asmakingsomeuse
of friction. Thebearing-typebolt can transfer larger forcethan thefriction boltsbut islessforgiving
with respect to theclearancespaceoften existingbetween thebolt and theplate. Theserequirethat
preciseholesbedrilled and at exact spacings. Theforcetransfer mechanismfor theseconnectionsis
shown in Figure10.9. In thebeam-to-column connection shown in Figure10.10, theboltsattached
to thecolumn aretension boltswhiletheboltson thebeamareslip-critical bolts.
Thetensionbolt transfersforceinthedirectionof thebolt axis. Thetensiontypeof bolt connection
iseasy to connect on site, but difcultiesarisein distributing forcesequally to each bolt, resulting
in reduced reliability. Tension bolts may also be used to connect box members of the towers of
suspension bridges where compression forces are larger than the tension forces. In this case, the
compression isshared with buttingsurfacesof theplatesand thetension iscarried by thebolts.
10.2.4 FabricationinShop
Steel bridgesarefabricated into membersin theshop yard and then transported to theconstruction
sitefor assembly. Ideally all constructional work would becompleted in theshop to get thehighest
quality in theminimum construction time. Thelarger and longer thememberscan be, thebetter,
within the restrictions set by transportation limits and erection tolerances. When crane ships for
erection and bargesfor transportation can beused, oneblock can weigh asmuch asathousand tons
c 1999by CRCPressLLC
FIGURE 10.9: Slip-critical and bearing-type connections. (From Nagai, N., BridgeEngineering,
Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1994. With permission.)
and beerected asawholeon thequay. In thesecasesthebridgeismadeof asinglecontinuousblock
and much of thehassleusually associated with assembly and erection isavoided.
10.2.5 ConstructiononSite
Thedesigner must consider theloadsthat occur duringconstruction, generally different fromthose
occurring after completion. Steel bridges are particularly prone to buckling during construction.
Theerectionplanmust bemadeprior tothemaindesignandmust becheckedfor everypossibleload
casethat may ariseduringerection, not only for strength but also for stability. Truck craneand bent
erection (or stagingerection); launchingerection; cableerection; cantilever erection; and largeblock
erection (or oating crane erection) are several techniques (see Figure 10.11). An example of the
largeblock erection isshown in Figure10.43, in which a186-m, 4500-ton center block istransported
by bargeand lifted.
10.2.6 Painting
Steel must bepainted to protect it from rusting. Thereisawidevariety of paints, and thelifeof a
steel structureislargely inuenced by itsquality. In areasnear thesea, thesalty air isparticularly
harmful toexposedsteel. Thecost of paintingishighbut isessential tothecontinuedgoodcondition
of thebridge. Thecolor of thepaint isalso an important consideration in termsof itspublicappeal
or aestheticquality.
c 1999by CRCPressLLC
FIGURE10.10: Tension-typeconnection.
10.3 ConcreteBridges
10.3.1 Introduction
For modern bridges, both structural concreteand steel givesatisfactory performance. Thechoice
betweenthetwomaterialsdependsmainlyuponthecost of constructionandmaintenance. Generally,
concretestructuresrequirelessmaintenancethan steel structures, but sincetherelativecost of steel
and concreteisdifferent from country to country, and may even vary throughout different partsof
thesamecountry, it isimpossibleto put onedenitively abovetheother in termsof economy.
In thissection, themain featuresof common typesof concretebridgesuperstructuresarebriey
discussed. Concretebridgesubstructureswill bediscussed in Section 10.4. A design exampleof a
two-span continuous, cast-in-place, prestressed concretebox girder bridgeisgiven in theAppendix.
For amoredetailed look at design proceduresfor concretebridges, referenceshould bemadeto the
recent booksof Gerwick [ 7] , Troitsky [ 24] , Xanthakos[ 26, 27] , and Tonias[ 23] .
10.3.2 ReinforcedConcreteBridges
Figure10.12 showsthetypical reinforced concretesectionscommonly used in highway bridgesu-
perstructures.
1. Slab
A reinforced concreteslab (Figure10.12a) isthemost economical bridgesuperstructure
for spansof up to approximately 40ft (12.2m). Theslab hassimpledetailsand standard
formwork and isneat, simple, and pleasingin appearance. Common spansrangefrom
16 to 44 ft (4.9 to 13.4 m) with structural depth-to-span ratiosof 0.06 for simplespans
and 0.045for continuousspans.
2. T-Beam(Deck Girder)
TheT-beams(Figure10.12b) aregenerallyeconomicfor spansof 40to60ft (12.2to18.3
m), but do requirecomplicated formwork, particularly for skewed bridges. Structural
c 1999by CRCPressLLC
FIGURE 10.11: Erections methods. (From Japan Construction Mechanization Association, Cost
Estimationof BridgeErection, Tokyo, Japan [ in Japanese] , 1991. With permission.)
depth-to-span ratios are 0.07 for simple spans and 0.065 for continuous spans. The
spacing of girders in a T-beam bridge depends on the overall width of the bridge, the
slab thickness, and thecost of theformwork and may betaken as1.5timesthestructural
depth. Themost commonly used spacingsarebetween 6and 10ft (1.8to 3.1m).
3. Cast-in-PlaceBox Girder
Box girdersliketheoneshown in Figure10.12c, areoften used for spansof 50 to 120 ft
c 1999by CRCPressLLC
FIGURE10.12: Typical reinforced concretesectionsin bridgesuperstructures.
(15.2to36.6m). Itsformwork for skewed structuresissimpler than that required for the
T-beam. Duetoexcessivedeadloaddeections, theuseof reinforcedconcreteboxgirders
over simplespansof 100ft (30.5m) or moremay not beeconomical. Thedepth-to-span
ratiosaretypically 0.06for simplespansand 0.055for continuousspanswith thegirders
spaced at 1.5 timesthestructural depth. Thehigh torsional resistanceof thebox girder
makesit particularlysuitablefor curved alignments, such astherampsontofreeways. Its
smooth owinglinesareappealingin metropolitan cities.
4. Design Consideration
A reinforced concretehighway bridgeshould bedesigned to satisfy thespecication or
coderequirements, suchastheAASHTO-LRFD[ 1] requirements(American Association
of StateHighway and Transportation OfcialsLoad and ResistanceFactor Design) for
all appropriateservice, fatigue, strength, andextremeevent limit states. In theAASHTO-
LRFD [ 1] , service limit states include cracking and deformation effects, and strength
limit statesconsider thestrength and stabilityof astructure. Abridgestructureisusually
designed for thestrength limit statesand isthen checked against theappropriateservice
and extremeevent limit states.
c 1999by CRCPressLLC
10.3.3 PrestressedConcreteBridges
Prestressed concrete, using high-strength materials, makes an attractive alternative for long-span
bridges. It hasbeen widely used in bridgestructuressincethe1950s.
1. Slab
Figure10.13showsFederal Highway Administration (FHWA) [6] standard typesof pre-
cast, prestressed, voided slabs and their sectional properties. While cast-in-place, pre-
stressed slab ismoreexpensivethan reinforced concreteslab, precast, prestressed slab is
economical when manyspansareinvolved. Common spansrangefrom20to50ft (6.1to
15.2m). Structural depth-to-span ratiosare0.03for both simpleand continuousspans.
FIGURE10.13: Federal Highway Administration (FHWA) precast, prestressed, voided slab sections.
(FromFederal Highway Administration, StandardPlansfor HighwayBridges, Vol. 1, ConcreteSuper-
structures, U.S. Department of Transportation, Washington, D.C., 1990. With permission.)
2. Precast I Girder
Figure10.14 showsAASHTO [ 6] standard typesof I-beams. Thesecompetewith steel
girders and generally cost morethan reinforced concretewith thesamedepth-to-span
ratios. Theformwork iscomplicated, particularly for skewed structures. Thesesections
areapplicableto spans30 to 120 ft (9.1 to 36.6 m). Structural depth-to-span ratiosare
0.055for simplespansand 0.05for continuousspans.
c 1999by CRCPressLLC
FIGURE10.14: Precast, prestressed AASHTO (American Association of StateHighway and Trans-
portation Ofcials) I-beam sections. (From Federal Highway Administration, StandardPlansfor
HighwayBridges, Vol. 1, ConcreteSuperstructures, U.S. Department of Transportation, Washington,
D.C., 1990. With permission.)
3. Box Girder
Figure 10.15 shows FHWA [ 6] standard types of precast box sections. The shape of a
cast-in-place, prestressed concrete box girder is similar to the conventional reinforced
concretebox girder (Figure10.12c). Thespacingof thegirderscan betaken astwicethe
structural depth. It isused mostly for spansof 100to600ft (30.5to182.9m). Structural
depth-to-span ratios are 0.045 for simple spans and 0.04 for continuous spans. These
c 1999by CRCPressLLC
sectionsareused frequently for simplespansof over 100ft (30.5m) and areparticularly
suitable for widening in order to control deections. About 70 to 80% of Californias
highway bridgesystemiscomposed of prestressed concretebox girder bridges.
FIGURE10.15: Federal HighwayAdministration(FHWA) precast, pretensionedboxsections. (From
Federal HighwayAdministration, StandardPlansforHighwayBridges,Vol. 1,ConcreteSuperstructures,
U.S. Department of Transportation, Washington, D.C., 1990. With permission.)
4. Segmental Bridge
Thesegmentally constructed bridgeshavebeen successfully developed by combiningthe
concepts of prestressing, box girder, and the cantilever construction [ 2, 20] . The rst
prestressedsegmental box girder bridgewasbuilt in Western Europein 1950. Californias
PineValley Bridge, asshown in Figure10.16 (composed of threespansof 340 ft [ 103.6
m] , 450 ft [ 137.2 m] , and 380 ft [ 115.8 ft] with thepier height of 340 ft [ 103.6 m] ), was
therst cast-in-placesegmental bridgebuilt in theU.S., in 1974.
Theprestressedsegmental bridgeswithprecast or cast-in-placesegmental canbeclassied
by theconstruction methods: (1) balanced cantilever, (2) span-by-span, (3) incremen-
tal launching, and (4) progressiveplacement. Theselection between cast-in-placeand
precast segmental, and among various construction methods, is dependent on project
features, siteconditions, environmental andpublicconstraints, construction timefor the
project, and equipment available. Table10.2 liststherangeof application of segmental
bridgesby span lengths[ 20] .
c 1999by CRCPressLLC
FIGURE10.16: a PineValley Bridge, California. Construction state. (FromCaliforniaDepartment of
Transportation. With permission.)
FIGURE10.16: b PineValley Bridge, California. Construction completed. (From CaliforniaDepart-
ment of Transportation. With permission.)
c 1999by CRCPressLLC
FIGURE10.17: A anged section at nominal moment capacity state.
TABLE10.2 Rangeof Application of Segmental BridgeTypeby Span Length
Span
ft (m) Bridgetypes
0150 (045.7) I-typepretensioned girder
100300 (30.591.4) Cast-in-placepost-tensioned box girder
100300 (30.591.4) Precast-balanced cantilever segmental, constant depth
200600 (61.0182.9) Precast-balanced cantilever segmental, variabledepth
2001000 (61.0304.8) Cast-in-placecantilever segmental
8001500 (243.8457.2) Cable-stay with balanced cantilever segmental
5. Design Consideration
Comparedtoreinforcedconcrete, themaindesignfeaturesof prestressedconcretearethat
stressesfor concreteandprestressingsteel anddeformationof structuresat eachstage(i.e.,
during construction, stressing, handling, transportation, and erection aswell asduring
theservicelife) andstressconcentrationsneedtobeinvestigated. Inthefollowing, weshall
briey discusstheAASHTO-LRFD [ 1] requirementsfor stresslimits, nominal exural
resistance, and shear resistancein designingaprestressed member.
a) StressLimits
Calculationsof stressesfor concreteand prestressingsteel arebased mainly on theelastic theory.
Tables10.3to 10.5list theAASHTO-LRFD [ 1] stresslimitsfor concreteand prestressingtendons.
b) Nominal Flexural Resistance, M
n
Flexural strengthisbasedontheassumptionsthat (1) thestrainislinearlydistributedacrossacross-
section (except for deep exural member); (2) themaximum usablestrain at extremecompressive
ber isequal to 0.003; (3) thetensilestrength of concreteisneglected; and (4) aconcretestressof
0.85f

c
isuniformly distributed over an equivalent compression zone. For amember with aanged
section (Figure10.17) subjected to uniaxial bending, theequationsof equilibriumareused to givea
nominal moment resistanceof:
M
n
= A
ps
f
ps
_
d
p

a
2
_
+A
s
f
y
(d
s

a
2
)
A

s
f

y
_
d

s

a
2
_
+0.85f

c
(b b
w
)
1
h
f
_
a
2

h
f
2
_
(10.1)
c 1999by CRCPressLLC
TABLE10.3 StressLimitsfor PrestressingTendons
Prestressingtendon type
Stress-relieved
strand and plain Deformed
Stress Prestressing high-strength LowRelaxation high-strength
type method bars strand bars
At jacking Pretensioning 0.72f
pu
0.78f
pu

(f
pj
) Post-tensioning 0.76f
pu
0.80f
pu
0.75f
pu
After Pretensioning 0.70f
pu
0.74f
pu

transfer Post-tensioning
(f
pt
) At anchorages
and couplers 0.70f
pu
0.70f
pu
0.66f
pu
immediately
after anchor set
General 0.70f
pu
0.74f
pu
0.66f
pu
At service After all losses
limit 0.80f
py
0.80f
py
0.80f
py
state(f
pe
)
FromAmericanAssociationof StateHighwayandTransportationOfcials, AASHTOLRFDBridge
DesignSpecications, First Edition, Washington, D.C., 1994. With permission.
a = c (10.2)
c =
A
ps
f
pu
+A
s
f
y
A

s
f

y
0.85
1
f

c
(b b
w
)h
f
0.85
1
f

c
b
w
+kA
ps
f
pu
d
p
(10.3)
f
ps
= f
pu
_
1 k
c
d
p
_
(10.4)
k = 2
_
1.04
f
py
f
pu
_
(10.5)
where A represents area; f is stress; b is the width of the compression face of member; b
w
is
the web width of a section; h
f
is the compression ange depth of a cross-section; d
p
and d
s
are
distancesfromextremecompression ber to thecentroid of prestressingtendonsand to centroid of
tension reinforcement, respectively; subscriptsc and y indicatespecied strength for concreteand
steel, respectively; subscriptsp and s signify prestressingsteel and reinforcement steel, respectively;
subscripts ps, py, and pu correspond to states of nominal moment capacity, yield, and specied
tensilestrength of prestressingsteel, respectively; superscript prime(

) representscompression; and

1
is the concrete stress block factor, equal to 0.85 f

c
4000 psi and 0.05 less for each 1000 psi
of f

c
in excessof 4000 psi, and minimum
1
= 0.65. Theaboveequationsalso can beused for a
rectangular section in which b
w
= b istaken.
Maximumreinforcement limit:
c
d
e
0.42 (10.6)
d
e
=
A
ps
f
ps
d
p
+A
s
f
y
d
s
A
ps
f
ps
+A
s
f
y
(10.7)
Minimumreinforcement limit:
M
n
1.2M
cr
(10.8)
in which istheexural resistancefactor 1.0for prestressedconcreteand0.9for reinforcedconcrete,
and M
cr
isthecrackingmoment strength given by theelasticstressdistribution and themodulusof
ruptureof concrete.
c 1999by CRCPressLLC
TABLE10.4 Temporary ConcreteStressLimitsat JackingStateBeforeLossesDueto Creep and
ShrinkageFully Prestressed Components
Stress Stress
type Areaand condition ksi (MPa)
Compressive Pretensioned 0.60f

ci
Post-tensioned 0.55f

ci
Precompressed tensilezonewithout bonded reinforcement N/A
Area other than the precompressed tensile zones and without
bonded auxiliary reinforcement
0.0948
_
f

ci
0.2
_
0.25
_
f

ci
1.38
_
Tensile
Nonsegmental
bridges
Areawithbondedreinforcement whichissufcient toresist 120%
of thetension forcein thecracked concretecomputed
0.22
_
f

ci
on thebasisof uncracked section
_
0.58
_
f

ci
_
Handlingstressesin prestressed piles
_
0.158
_
f

ci
_
_
0.415
_
f

ci
_
TypeAjointswithminimumbondedaux-
iliary reinforcement through the
0.0948
_
f

ci
max. tension
Longitudinal stress
through joint in
precompressed
jointswhich issufcient to carry thecal-
culated tensileforceat astressof 0.5 f
y
with internal tendons
(0.25
_
f

ci
max. tension)
tensilezone Type A joints without the minimum
bonded auxiliary reinforcement through
thejointswith internal tendons
No tension
TypeBwith external tendons 0.2min. compression
(1.38min. compression)
Segmental Transversestress For any typeof joint 0.0948
_
f

c
max. tension
bridges through joints (0.25
_
f

c
max. tension)
Without bonded non-prestressed rein-
forcement
No tension
Other area Bonded reinforcement is sufcient to
carry thecalculated tensileforcein the
0.19
_
f

ci
concrete on the assumption of an un-
cracked section at astressof 0.5f
sy
(0.50
_
f

ci
)
Note: TypeA jointsarecast-in-placejointsof wet concreteand/or epoxy between precast units. TypeB jointsaredry joints
between precast units.
From American Association of StateHighway and Transportation Ofcials, AASHTOLRFDBridgeDesignSpecications, First
Edition, Washington, D.C., 1994. With permission.
c) Nominal Shear Resistance, V
n
Thenominal shear resistanceshall bedetermined by thefollowingformulas:
V
n
= thelesser of
_
V
c
+V
s
+V
p
0.25f

c
b

+V
p
(10.9)
where
V
c
=
_
0.0316
_
f

c
b

(ksi)
0.083
_
f

c
b

(MPa)
(10.10)
V
s
=
A

f
y
d

(cos +cos ) sin

s
(10.11)
where b

is the effective web width determined by subtracting the diameters of ungrouted ducts
or one-half the diameters of grouted ducts; d

is the effective depth between the resultants of the

tensile and compressive forces due to exure, but not less than the greater of 0.9 d
e
or 0.72h; A

is the area of transverse reinforcement within distance s; s is the spacing of the stirrups; is the
angleof inclination of transversereinforcement to thelongitudinal axis; isafactor indicatingthe
c 1999by CRCPressLLC
TABLE10.5 ConcreteStressLimitsat ServiceLimit StateAfter All LossesFully Prestressed
Components
Stress Stress
type Areaand condition ksi (MPa)
Nonsegmental bridge at servicestate 0.45f

c
Compressive Nonsegmental bridgeduringshippingand handling 0.60f

c
Segmental bridgeduringshippingand handling 0.45f

c
With bonded prestressingtendons 0.19
_
f

c
other than piles (0.50
_
f

c
)
Precompressed Subjected to severecorrosive
Tensile tensilezoneassuming
uncracked sections
conditions 0.0948
_
f

c
Nonsegmental
_
0.25
_
f

c
_
bridges With unbonded prestressingtendon No tension
TypeAjointswithminimumbondedaux-
iliary reinforcement through the joints
which issufcient tocarrythecalculated
tensileforceat astressof 0.5f
y
with in-
ternal tendons
0.0948
_
f

c
(0.25
_
f

c
)
Longitudinal stressin
precompressed tensile
zone
Type A joints without the minimum
bonded auxiliary reinforcement through
thejoints
No tension
TypeBwith external tendons 0.2min. compression
(1.38min. compression)
Segmental
bridges
Transversestressin
precompressed tensile
zone
For any typeof joint 0.0948
_
f

c
_
0.25
_
f

c
_
TypeA joint without minimumbonded
auxiliary reinforcement through joints
No tension
Other area
(without bonded
reinforcement)
Bonded reinforcement is sufcient to
carry the calculated tensile force in the
concrete on the assumption of an un-
cracked section at astressof 0.5f
sy
0.19
_
f

c
_
0.50
_
f

c
_
Note: TypeA jointsarecast-in-placejointsof wet concreteand/or epoxy between precast units. TypeBjointsaredry joints
between precast units.
FromAmerican Association of StateHighwayand Transportation Ofcials, AASHTOLRFDBridgeDesignSpecications, First
Edition, Washington, D.C., 1994. With permission.
abilityof diagonallycrackedconcretetotransmit tension; and istheangleof inclinationof diagonal
compressivestresses(Figure10.18). Thevaluesof and for sectionswithtransversereinforcement
aregiven in Table10.6. In thistable, theshear stress, , and strain,
x
, in thereinforcement on the
exural tension sideof themember aredetermined by:
=
V
u
V
p
b

(10.12)

x
=
M
u
d

+0.5N
u
+0.5V
u
cot A
ps
f
po
E
s
A
s
+E
p
A
ps
0.002 (10.13)
where M
u
and N
u
are the factored moment and axial force (taken as positive if compressive),
respectively, associated with V
u
, and f
po
is the stress in prestressing steel when the stress in the
surroundingconcreteiszero and can beconservatively taken astheeffectivestressafter losses, f
pe
.
Whenthevalueof
x
calculatedfromtheaboveequationisnegative, itsabsolutevalueshall bereduced
c 1999by CRCPressLLC
FIGURE10.18: Illustrationof A
c
for shear strengthcalculation. (FromAmericanAssociationof State
Highway and Transportation Ofcials, AASHTO LRFD BridgeDesign Specications, First Edition,
Washington, D.C., 1994. With permission.)
TABLE10.6 Valuesof and for Sectionswith TransverseReinforcement
Angle
x
1000

c
(degree) 0.2 0.15 0.1 0 0.125 0.25 0.50 0.75 1.00 1.50 2.00
0.05 27.0 27.0 27.0 27.0 27.0 28.5 29.0 33.0 36.0 41.0 43.0
6.78 6.17 5.63 4.88 3.99 3.49 2.51 2.37 2.23 1.95 1.72
0.075 27.0 27.0 27.0 27.0 27.0 27.5 30.0 33.5 36.0 40.0 42.0
6.78 6.17 5.63 4.88 3.65 3.01 2.47 2.33 2.16 1.90 1.65
0.100 23.5 23.5 23.5 23.5 24.0 26.5 30.5 34.0 36.0 38.0 39.0
6.50 5.87 5.31 3.26 2.61 2.54 2.41 2.28 2.09 1.72 1.45
0.125 20.0 21.0 22.0 23.5 26.0 28.0 31.5 34.0 36.0 37.0 38.0
2.71 2.71 2.71 2.60 2.57 2.50 2.37 2.18 2.01 1.60 1.35
0.150 22.0 22.5 23.5 25.0 27.0 29.0 32.0 34.0 36.0 36.5 37.0
2.66 2.61 2.61 2.55 2.50 2.45 2.28 2.06 1.93 1.50 1.24
0.175 23.5 24.0 25.0 26.5 28.0 30.0 32.5 34.0 35.0 35.5 36.0
2.59 2.58 2.54 2.50 2.41 2.39 2.20 1.95 1.74 1.35 1.11
0.200 25.0 25.5 26.5 27.5 29.0 31.0 33.0 34.0 34.5 35.0 36.0
2.55 2.49 2.48 2.45 2.37 2.33 2.10 1.82 1.58 1.21 1.00
0.225 26.5 27.0 27.5 29.0 30.5 32.0 33.0 34.0 34.5 36.5 39.0
2.45 2.38 2.43 2.37 2.33 2.27 1.92 1.67 1.43 1.18 1.14
0.250 28.0 28.5 29.0 30.0 31.0 32.0 33.0 34.0 35.5 38.5 41.5
2.36 2.32 2.36 2.30 2.28 2.01 1.64 1.52 1.40 1.30 1.25
From American Association of StateHighway and Transportation Ofcials, AASHTOLRFDBridgeDesignSpecications,
First Edition, Washington, D.C., 1994. With permission.
by multiplyingby thefactor F

, taken as:
F

=
E
s
A
s
+E
p
A
ps
E
c
A
c
+E
s
A
s
+E
p
A
ps
(10.14)
whereE
s
, E
p
, and E
c
aremodulesof elasticity for reinforcement, prestressing steel, and concrete,
respectively, and A
c
istheareaof concreteon theexural tension sideof themember, asshown in
Figure10.18.
Minimumtransversereinforcement:
A
min
=
_
0.0316
_
f

c
b

S
f
y
(ksi)
0.083
_
f

c
b

S
f
y
(MPa)
(10.15)
c 1999by CRCPressLLC
Maximumspacingof transversereinforcement:
For V
u
< 0.1f

c
b

s
max
= thesmaller of
_
0.8d

24 in. (600 mm)

(10.16)
For V
u
0.1f

c
b

s
max
= thesmaller of
_
0.4d

12 in. (300 mm)

(10.17)
10.4 ConcreteSubstructures
10.4.1 Introduction
Bridgesubstructurestransfer trafc loadsfrom thesuperstructureto thefootingsand foundations.
Vertical intermediatesupports(piersor bents) and end supports(abutments) areincluded.
10.4.2 BentsandPiers
1. PileBents
Pileextension, asshowninFigure10.19a, isusedfor slabandT-beambridges. It isusually
used to crossstreamswhen debrisisnot aproblem.
FIGURE10.19: Bridgesubstructurespiersandbents. (FromCaliforniaDepartment of Transporta-
tion, BridgeDesignAidsManual, Sacramento, CA, 1990. With permission.)
2. Solid Piers
Figure10.19b showsatypical solid pier, used mostly when streamdebrisor fast currents
arepresent. Theseareused for long spansand can besupported by spread footingsor
pilefoundations.
3. Column Bents
c 1999by CRCPressLLC
Columnbents(Figure10.19c) aregenerallyusedondrylandstructuresandaresupported
by spread footingsor pilefoundations. Multi-column bentsaredesirablefor bridgesin
seismiczones. Thesingle-columnbent, suchasaTbent (Figure10.19d), modiedTbent
(Cbent) (Figure10.19e), or outrigger bent (Figure10.19f ), maybeusedwhenthelocation
of thecolumnsisrestricted and changesof thealignment areimpossible. To achievea
pleasing appearanceat theminimum cost using standard column shapes, Caltrans [ 3]
developedstandardarchitectural columns (Figure10.20). Prismaticsectionsof column
types1and 1W, with one-way aresof column types2and 2W, and with two-way ares
of column types3and 3W may beused for varioushighway bridges.
10.4.3 Abutments
Abutments are the end supports of a bridge. Figure 10.21 shows the typical abutments used for
highway bridges. Theseven typesof abutmentscan bedivided into two categories: open and closed
ends. Selection of an abutment typedependson therequirementsfor structural support, movement,
drainage, road approach, and earthquakes.
1. Open-End Abutments
Open-endabutmentsincludediaphragmabutmentsandshort-seat abutments. Theseare
themost frequentlyusedabutmentsandareusuallythemost economical, adaptable, and
attractive. Thebasic structural differencebetween thetwo typesisthat seat abutments
permit thesuperstructuretomoveindependentlyfromtheabutment whilethediaphragm
abutment does not. Since open-end abutments have lower abutment walls, there is
lesssettlement in theroad approachesthan that experienced by higher backlled closed
abutments. They also providefor moreeconomical wideningthan closed abutments.
2. Closed-End Abutments
Closed-end abutments include cantilever, strutted, rigid frame, bin, and closure abut-
ments. Thesearelesscommonlyused, but for bridgewideningsof thesamekind, unusual
sites, or in tightlyconstrainedurban locations. Rigidframeabutmentsaregenerallyused
with tunnel-typesingle-span connectorsand overhead structureswhich permit passage
through aroadway embankment. Becausethestructural supportsareadjacent to trafc
thesehaveahigh initial cost and present aclosed appearanceto approachingtrafc.
10.4.4 DesignConsideration
After therecent 1989 Loma Prieta and the1994 NorthridgeEarthquakes in theU.S. and the1995
Kobeearthquakein Japan, major damageswerefound in substructures. Special attention, therefore,
must bepaid to seismic effectsand thedetailingof theductilestructures. Boundary conditionsand
soilfoundationstructureinteraction in seismicanalysesshould also becarefully considered.
10.5 Floor System
10.5.1 Introduction
Theoor systemof abridgeusually consistsof adeck, oor beams, and stringers. Thedeck directly
supportstheliveload. Floor beamsaswell asstringers, shown in Figure10.22, form agrillageand
transmit the load from the deck to the main girders. The oor beams and stringers are used for
framed bridges, i.e., truss, rahmen, and arch bridges(seeFigures10.40, 10.45, and 10.47), in which
thespacing of themain girdersor trussesislarge. In an upper deck typeof plategirder bridgethe
c 1999by CRCPressLLC
FIGURE10.20: Caltrans(CaliforniaDepartment of Transportation) standardarchitectural columns.
(FromCaliforniaDepartment of Transportation, BridgeDesignAidsManual, Sacramento, CA, 1990.
With permission.)
deck isdirectly supported by themain girders, and often thereisno oor system becausethemain
girdersrun in parallel and closetogether.
Theoor systemisclassiedassuitablefor either highwayor railroadbridges. Thedeckof ahighway
bridgeisdesigned for thewheel loadsof trucksusingplatebendingtheoryin twodimensions. Often
in design practice, however, thisplatetheory isreduced toequivalent one-dimensional beamtheory.
Thematerialsused arealso classied into concrete, steel, or wood.
Therecent inux of trafc ow hasseverely fatigued existing oor systems. Cracksin concrete
c 1999by CRCPressLLC
FIGURE10.21: Typical typesof abutments. (FromCaliforniaDepartment of Transportation, Bridge
DesignAidsManual, Sacramento, CA, 1990. With permission.)
decksand connectionsof oor system areoften found in old bridgesthat havebeen in servicefor
many years.
10.5.2 Decks
1. ConcreteDeck
A reinforced concrete deck slab is most commonly used in highway bridges. It is the
deck that ismost susceptibleto damagecaused by theowof trafc, which continuesto
increase. Urban highwaysareexposed to heavy trafcand must berepaired frequently.
Recently, a composite deck slab was developed to increase the strength, ductility, and
durability of deckswithout increasingtheir weight or affectingthecost and duration of
construction. In a composite slab, the bottom steel plate serves both as a part of the
slab and the formwork for pouring the concrete. There are many ways of combining
thesteel plateand thereinforcement. A typical exampleisshown in Figure10.23. This
slab is prefabricated in the yard and then the concrete is poured on site after girders
c 1999by CRCPressLLC
FIGURE10.22: Floor system. (FromNagai, N., BridgeEngineering, Kyoritsu PublishingCo., Tokyo,
Japan [ in Japanese] , 1994. With permission.)
havebeen placed. A precast, prestressed deck may reducethetimerequired to complete
construction.
FIGURE10.23: Compositedeck. (FromJapan Association of Steel BridgeConstruction, Planningof
Steel Bridges, Tokyo [ in Japanese] , 1988. With permission.)
2. Steel Deck
For longspans, thesteel deck isused to minimizetheweight of thedeck. Thesteel deck
plate is stiffened with longitudinal and transverse ribs as shown in Figure 10.24. The
steel deck also works as the upper ange of the supporting girders. The pavement on
c 1999by CRCPressLLC
thesteel deck should becarefully nished to prevent water frompenetratingthrough the
pavement and causingthesteel deck to rust.
FIGURE10.24: Steel platedeck. (From Japan Association of Steel BridgeConstruction, Outlineof
Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
10.5.3 Pavement
Thepavement on thedeck providesasmooth drivingsurfaceand preventsrain water from seeping
into thereinforcingbarsand steel deck below. Alayer of waterproongmay beinserted between the
pavement and thedeck. Asphalt ismost commonly used to pavehighway bridges. Itsthicknessis
usually 5to 10cmon highwaysand 2to 3cmon pedestrian bridges.
10.5.4 Stringers
The stringers support the deck directly and transmit the loads to oor beams, as can be seen in
Figure10.22. They areplaced in thelongitudinal direction just likethemain girdersarein aplate
girder bridgeand thusprovidemuch thesamekind of support.
Thestringersmust besufcientlystiff in bendingtoprevent cracksfromformingin thedeck or on
thepavement surface. Thedesign codesusually limit thevertical displacement caused by theweight
of atruck.
10.5.5 Floor Beams
Theoor beamsareplaced in thetransversedirection and connected by high-tension boltsto the
trussframeor arch, asshown in Figure10.22. Theoor beamssupport thestringersand transmit
theloadstomain girders, trusses, or arches. In other words, themain trussor arch receivestheloads
indirectly viatheoor beams. Theoor beamsalso providetransversestiffnessto bridgesand thus
improvetheoverall torsional resistance.
c 1999by CRCPressLLC
10.6 Bearings, ExpansionJoints, andRailings
10.6.1 Introduction
Aside from the main components, such as the girders or the oor structure, other parts such as
bearings(shoes), expansion joints, guardrailings, drainagepaths, lighting, andsound-proongwalls
also makeup thestructureof abridge. Each playsaminor part but providesan essential function.
Drainsush rain water off and wash away dust. Guardrailingsand lightsadd to theaestheticquality
of thedesign aswell asprovidingtheir obviousoriginal functions. A sound-proongwall may take
awayfromthebeautyof thestructurebut might berequiredbylawinurbanareastoisolatethesound
of trafc from thesurroundingresidents. In thefollowingsection, bearings, expansion joints, and
guardrailingsarediscussed.
10.6.2 Bearings(Shoes)
Bearingssupport thesuperstructure(themain girders, trusses, or arches) and transmit theloadsto
the substructure (abutments or piers). The bearings connect the upper and lower structures and
carrythewholeweight of thesuperstructure. Thebearingsaredesignedtoresist thesereaction forces
by providingsupport conditionsthat arexed or hinged. Thehinged bearingsmay bemovableor
immovable; horizontal movement isrestrained or unrestrained, i.e., horizontal reaction isproduced
or not. Theamount of thehorizontal movement isdetermined by calculatingtheelongation dueto
atemperaturechange.
Many bearingswerefound to havesustained extensivedamageduringthe1995KobeEarthquake
in Japan, dueto stressconcentrations, which aretheweak spotsalongthebridge. Thebearingsmay
play theroleof afuseto keep damagefrom occurringat vital sectionsof thebridge, but therisk of
thesuperstructurefalling down goes up. Thegirder-to-girder or girder-to-abutment connections
prevent thegirdersfromcollapsingduringstrongearthquakes.
Many typesof bearingsareavailable. Someareshown in Figure10.25and briey explained in the
following:
Linebearings: The contacting line between the upper plate and the bottom round surface
providesrotational capability aswell assliding. Theseareused in small bridges.
Platebearings: Thebearingplatehasaplanesurfaceon thetop sidewhich allowsslidingand a
spherical surfaceon thebottomallowingrotation. Theplateisplaced between theupper
and lower shoes.
Hinged bearings(pin bearings): Apin isinserted between theupper and lower shoesallowing
rotation but no translation in longitudinal direction.
Roller bearings: Lateral translationisunrestrainedbyusingsingleor multiplerollersfor hinged
bearingsor spherical bearings.
Spherical bearings(pivot bearings): Convex and concave spherical surfaces allow rotation in
all directions and no lateral movement. The two types are: a point contact for large
differencesin theradii of each sphereand asurfacecontact for small differencesin their
radii.
Pendel bearings: An eyebar connectsthesuperstructureand thesubstructureby apin at each
end. Longitudinal movement ispermittedbyincliningtheeyebar; therefore, thedistance
of thepinsat endsshould beproperly determined. Theseareused to provideanegative
reaction in cable-stayed bridges. Thereisno resistancein thetransversedirection.
Wind bearings: Thistypeof bearingprovidestransverseresistancefor wind and isoften used
with pendel bearings.
c 1999by CRCPressLLC
FIGURE10.25: Typesof bearings. (From Japan Association of Steel BridgeConstruction, A Guide
Bookof BearingDesignfor Steel Bridges, Tokyo [ in Japanese] , 1984. With permission.)
Elastomericbearings: Theexibilityof elastomericor leadrubber bearingsallowsbothrotation
and horizontal movement. Figure10.26 explainsaprincipleof rubber-layered bearings
by comparing with a unit rubber. A layered rubber is stiff, unlike a unit rubber, for
vertical compressionbecausethesteel platesplacedbetweentherubber restrainthevertical
deformation of therubber, but exiblefor horizontal shear forcelikeaunit rubber. The
exibility absorbs horizontal seismic energy and is ideally suited to resist earthquake
actions. Since the disaster of the 1995 Kobe Earthquake in Japan, elastomeric rubber
bearingshavebecomemoreandmorepopular, but whether theyeffectivelysustainsevere
vertical actionswithout damageisnot certied.
Oil damper bearings: Theoil damper bearingsmoveunder slowactions(such astemperature
changes) but do not move under quick movements (such as those of an earthquake).
They areused in continuousspan bridgesto distributeseismicforces.
c 1999by CRCPressLLC
FIGURE10.26: Propertiesof elastomericbearings.
A selection from these types of bearings is made according to the size of the bridge and the
magnitudeof predicted downward or upward reaction forces.
10.6.3 ExpansionJoints
Expansion joints areprovided to allow a bridgeto adjust its length under changes in temperature
or deformation by external loads. They aredesigned accordingto expandinglength and material as
classied in Figure10.27. Steel expansion jointsaremost commonly used. A defect isoften found
at theboundary between thesteel and theconcreteslab wherethedisturbingjolt isgiven to drivers
c 1999by CRCPressLLC
as they pass over the junction. To solve this problem, rubber joints are used on the road surface
to provideasmooth transition for modern bridgeconstruction (seeFigure10.27e), or continuous
girdersaremorecommonly adopted than simplegirders.
FIGURE10.27: Typesof expansion joints. (FromJapan Association of Steel BridgeConstruction, A
GuideBookof ExpansionJointDesignfor Steel Bridges, Tokyo[ in Japanese] , 1984. With permission.)
10.6.4 Railings
Guardrailingsareprovidedtoensurevehiclesandpedestriansdonot fall off thebridge. Theymaybe
ahandrail for pedestrians, aheavier guardfor vehicles, or acommonrailingfor both. Thesearemade
frommaterialssuch asconcrete, steel, or aluminum. Theguardrailingsarelocated prominently and
arethusopen to thecritical eyeof thepublic. It isimportant that they not only keep trafc within
boundariesbut also add to theaestheticappeal of thewholebridge(Figure10.28).
c 1999by CRCPressLLC
FIGURE10.28: Pedestrian railing. (FromJapan Association of Steel BridgeConstruction, Outlineof
Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
10.7 Girder Bridges
10.7.1 Structural Features
Girder bridges are structurally the simplest and the most common. They consist of a oor slab,
girders, andthebearingswhichsupport andtransmit gravityloadstothesubstructure. Girdersresist
bending momentsand shear forcesand areused to span short distances. Girdersareclassied by
material into steel plateand box girders, reinforced or prestressed concreteT-beams, and composite
girders. The box girder is also used often for prestressed concrete continuous bridges. The steel
girder bridgesareexplained in thissection; theconcretebridgesweredescribed in Section 10.3.
Figure10.29showsthestructural compositionof plateandboxgirder bridgesandtheloadtransfer
path. In plate girder bridges, the live load is directly supported by the slab and then by the main
girders. In box girder bridgestheforcesaretaken rst by theslab, then supported by thestringers
c 1999by CRCPressLLC
and oor beamsin conjunction with themain box girders, and nally taken to thesubstructureand
foundation through thebearings.
FIGURE10.29: Steel girder bridges. (From Nagai, N., BridgeEngineering, Kyoritsu PublishingCo.,
Tokyo, Japan [ in Japanese] , 1994. With permission.)
Girdersareclassiedasnoncompositeor composite, that is, whether thesteel girdersact intandem
with theconcreteslab (usingshear connectors) or not. Sincecompositegirdersmakeuseof thebest
propertiesof bothsteel andconcrete, theyareoftentherational andeconomicchoice. Lessfrequently
H or I shapesareused for themain girdersin short-span noncompositebridges.
10.7.2 PlateGirder (Noncomposite)
Theplategirder isthemost economical shapedesigned to resist bendingand shear; themoment of
inertiaisgreatest for arelatively low weight per unit length. Figure10.30 showsaplan of atypical
plategirder bridgewith four main girdersspanning30mand awidth of 8.5m.
Thegravityloadsaresupportedbyseveral main plategirders, eachmanufacturedbyweldingthree
c 1999by CRCPressLLC
FIGURE10.30: General plansof atypical plategirder bridge. (From Tachibana, Y. and Nakai, H.,
BridgeEngineering, Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
plates: an upper and lower ange and a web. Figure 10.31 shows a block of plate girder and its
fabrication process. The web and the anges are cut from steel plate and welded. The block is
fabricated in theshop and transported to theconstruction sitefor erection.
Thedesign procedurefor plategirders, primarily thesizingof thethreeplates, isasfollows:
1. Web height: Theweb height isthefundamental design factor affecting theweight and
cost of thebridge. If theheight istoo small, theangesneed to belargeand thedead
weight increases. Theheight (h) isdetermined empirically by dividing thespan length
(L) byareasonable factor. Common ratiosareh/L =1/18to1/20for highwaybridges
and alittlesmaller for railway bridges. Theweb height also inuencesthestiffnessof the
c 1999by CRCPressLLC
FIGURE10.31: Fabrication of plategirder block.
bridge. Greater heightsgenerally producegreater stiffness. However, if theheight istoo
great, theweb becomesunstableand must haveitsthicknesssupplemented or stiffeners
added. Thesemeasuresincreasetheweight and thecost. In addition, plategirderswith
excessively deep web and small angesareliableto bucklelaterally.
2. Web thickness: Theweb primarily resistsshear forces, which arenot usually signicant
when the web height is properly designed. The shear force is generally assumed to be
distributed uniformly acrosstheweb instead of usingtheexact equation of beamtheory.
The web thickness (t ) is determined such that thinner is better as long as buckling is
prevented. Sincetheweb doesnot contributemuch to thebendingresistance, thin webs
are most economical but the possibility of buckling increases. Therefore, the web is
usually stiffened by horizontal and vertical stiffeners, which will bediscussed later (see
Figure10.34). It isnot primarily strength but rather stiffnessthat controlsthedesign of
webs.
3. Areaof anges: After thesizesof webaredetermined, theangesaredesigned. Theanges
work mostly in bendingand therequired areaiscalculated usingequilibriumconditions
imposed on theinternal and external bending moment. A selection of strength for the
steel material isprincipally madeat thisstagein thedesign process.
4. Width and thicknessof anges: Thewidth and thicknesscan bedetermined by ensur-
ing that the area of the anges falls under the limiting width-to-thickness ratio, b/t
(Figure10.32), asspecied in design codes. If theangesaretoo thin (i.e., thewidth-to-
thicknessratioistoolarge), thecompression angemaybuckleor thetension angemay
bedistorted by theheat of welding. Thus, thethicknessof both angesmust bechecked.
Sinceplategirdershavelittletorsional resistance, special attention should bepaid to lat-
eral torsional buckling. To prevent thisphenomenon, thecompression angemust have
sufcient width toresist out-of-plane bending. Figure10.33showsthelateral torsional
bucklingthat may occur by bendingwith respect to strongaxis.
After determining themember sizes, calculations of theresisting moment capacity aremadeto
ensure code requirements are satised. If these fail, the above steps must be repeated until the
specicationsaremet.
A fewother important factorsin thedesign of girder bridgeswill beexplained in thefollowing:
Design of web stiffeners: Thehorizontal and vertical stiffenersshould beattached to theweb
c 1999by CRCPressLLC
FIGURE10.32: Local bucklingof compression ange.
FIGURE10.33: Lateral torsional buckling.
(Figure 10.34) when it is relatively thin. Bending moment produces compression and
tension in theweb, separatedbyaneutral axis. Thehorizontal stiffener preventsbuckling
due to bending and is therefore attached to the compression side (the top half for a
simply supported girder). Sincethebending moment is largest near themidspan of a
simply supported girder, the horizontal stiffeners are usually located there. If the web
is not too deep nor its thickness too small, no stiffeners are necessary and fabrication
costsarereduced. Vertical stiffeners, on theother hand, prevent shear buckling, which is
producedbythetension andcompression eldsin diagonal directions. Thecompression
eld causesshear buckling. Sincetheshear forceislargest near thesupports, themost
vertical stiffenersareneeded there. Bearingstiffeners, which aredesigned independently
just as any other compression member would be, are also required at the supports to
combat largereaction forces. Bucklingpatternsof aweb areshown in Figure10.34.
Variablesections: Thevariablecross-sectionsmay beused to savematerial and cost wherethe
bendingmoment issmaller, that is, near theendof thespan (seeFigure10.31). However,
thisreduction increasesthelabor required for weldingand fabrication. Thecost of labor
andmaterial must bebalancedandtradedoff. In todaysindustrial climate, labor ismore
important andcostlythanthematerial. Therefore, thechangeof girder sectionisavoided.
Likewise, thick plates are often specied to eliminate the number of stiffeners needed,
thusto reducethenecessary labor.
c 1999by CRCPressLLC
FIGURE10.34: Bucklingand stiffenersof web.
10.7.3 CompositeGirder
If two beamsaresimply laid oneupon theother, asshown in Figure10.35a, they act separately and
only sharetheload depending on their relativeexural stiffness. In thiscase, slip occursalong the
boundarybetween thebeams. However, if thetwobeamsareconnected and slip prevented asshown
in Figure 10.35b, they act as a unit, i.e., a composite girder. For composite plate girder bridges,
thesteel girder and theconcreteslab arejoined by shear connectors. In thisway, theconcreteslab
becomesintegral withthegirder andusuallytakesmost of thecompressioncomponent of thebending
moment whilethesteel plategirder takesthetension. Compositegirdersaremuch moreeffective
than thesimply tiered girder.
FIGURE10.35: Principleof tiered beam and compositebeam. (From Tachibana, Y. and Nakai, H.,
BridgeEngineering, Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
Let usconsider thetwo casesshown in Figure10.35and notethedifferencebetween tiered beams
and compositebeams. Both havethesamecross-sectionsand aresubjected toaconcentrated load at
midspan. Themoment of inertiafor thecompositebeamisfour timesthat of thetiered beams, thus
c 1999by CRCPressLLC
theresultingvertical deection isone-fourth. Themaximum bendingstressin theextreme(top or
bottom) ber ishalf that of thetiered beamconguration.
The corresponding stress distributions are shown in Figure 10.36. Points S and V are the
center of areaof thesteel section and thecompositesection, respectively. Accordingto beamtheory,
thestrain distribution islinear but thestressdistribution hasastep changeat theboundary between
thesteel and concrete.
FIGURE10.36: Section of compositegirder. (FromTachibana, Y. and Nakai, H., BridgeEngineering,
Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
Threetypesof shear connectorsstuds, horseshoes, andsteel blocksareshown in Figure10.37.
Studsaremost commonly used sincethey areeasily welded to thecompression angeby theelectric
FIGURE10.37: Typesof shear connectors. (FromNagai, N., BridgeEngineering, Kyoritsu Publishing
Co., Tokyo, Japan [ in Japanese] , 1994. With permission.)
resistancewelding, but theweld inspection isa cumbersometask. If theweld on a certain stud is
poor, thestudmayshear off andtrigger atotallyunforeseenfailuremode. Other typesareconsidered
to maintain morereliability.
Shear connectorsareneeded most near theendsof thespan, wheretheshear forceislargest. This
region isillustrated in Figure10.35a, which showsthemaximum shift dueto slip occursat theends
of tiered beams. It isthisslip that isrestrained by theshear connectors.
c 1999by CRCPressLLC
10.7.4 GrillageGirder
When girders are placed in a row and connected transversely by oor beams, the truck loads are
distributed by theoor beamsto thegirders. Thissystem iscalled agrillageof girders. If themain
girdersareplategirders, no stiffnessin torsion isconsidered. On theother hand, box girdersand
concretegirderscan beanalyzedassumingstiffnessisavailabletoresist torsion. Floor beamsincrease
thetorsional resistanceof thewholestructural systemof thebridge.
Let usconsider thestructural systemshown in Figure10.38ato observetheload distribution in a
grillagesystem. Thisgrillagehasthreegirderswith oneoor beam at midspan. In thiscase, there
FIGURE10.38: Grillagegirders. (From Tachibana, Y. and Nakai, H., BridgeEngineering, Kyoritsu
PublishingCo., Tokyo, Japan [ in Japanese] , 1996. With permission.)
arethreenodal forcesat theintersectionsof thegirdersand theoor beambut only two equilibrium
equations(V = 0andM = 0). Thus, it becomesonedegreestaticallyindeterminate. If wedisconnect
theintersection between main girder B and theoor beamand apply apair of indeterminateforces,
X, at point b, asshowninFigure10.38b, Xcanbeobtainedusingthecompatibilityconditionat point
c 1999by CRCPressLLC
b. Oncetheforce, X, isfound, thesectional forcesin thegirderscan becalculated. Thisstructural
systemiscommonly applied to thepractical design of plategirder bridges.
10.7.5 BoxGirder
Structural conguration of box girdersisillustrated in Figure10.39. Sincethebox girder isaclosed
section, its resistance to torsion is high with no loss of strength in bending and shear. On the
other hand, plategirdersareopen sectionsgenerally only considered effectivein resisting bending
and shear. Steel plates with longitudinal and transversestiffeners areoften used for decks on box
girder or thin-walled structuresinstead of aconcreteslab (Figure10.39b) although aconcreteslab is
permissible.
FIGURE10.39: Box girders. (From Nagai, N., BridgeEngineering, Kyoritsu PublishingCo., Tokyo,
Japan [ in Japanese] , 1994. With permission.)
Torsion isresisted in two parts: puretorsion (St. Venant torsion) and warpingtorsion. Thepure
torsional resistanceof I-plategirdersisnegligible. However, for closed sectionssuch asabox girder,
thepuretorsional resistanceisconsiderable, makingthem particularly suited for curved bridgesor
long-span bridges. On the other hand, the warping torsion for box sections is negligible. The I-
section girder hassomewarpingresistancebut it isnot largecompared to thepuretorsion of closed
sections.
10.8 TrussBridges
10.8.1 Structural Features
Thestructural layout of atrussbridgeisshown in Figure10.40 for athrough bridgewith thedeck
locatedat thelevel of lower chords. Theoor slab, whichcarriestheliveload, issupportedbytheoor
systemof stringersandcrossbeams. Theloadistransmittedtothemaintrussesat nodal connections,
oneon each sideof thebridge, through theoor system and nally to thebearings. Lateral braces,
which also areatrussframe, areattached to theupper and lower chordsto resist horizontal forces
such aswind and earthquakeloadsaswell astorsional moments. Theportal frameat theentrance
providestransition of horizontal forcesfromtheupper chordsto thesubstructure.
Truss bridges can take the form of a deck bridge as well as a through bridge. In this case, the
concreteslab ismounted on theupper chordsand thesway bracing isplaced between thevertical
membersof two main trussesto providelateral stability.
c 1999by CRCPressLLC
FIGURE10.40: Trussbridge. (From Nagai, N., BridgeEngineering, Kyoritsu PublishingCo., Tokyo,
Japan [ in Japanese] , 1994. With permission.)
A trussiscomposed of upper and lower chords, joined by diagonal and vertical members(web
members). Thisframeactioncorrespondstobeamactioninthat theupper andlower chordsperform
likeangesand thediagonal bracesbehavein much thesameway astheweb plate. Thechordsare
mainly in chargeof bending moment whiletheweb memberstaketheshear force. Trussesarean
assemblyof bars, not plates, andthusarecomparativelyeasier toerect on siteandareoften thechoice
for longbridges.
10.8.2 Typesof Trusses
Figure10.41showssometypical trusses. AWarrentrussisthemost commonandisaframecomposed
of isoscelestriangles, wherethewebmembersareeither incompressionor tension. Thewebmembers
of a Pratt truss are vertical and diagonal members where the diagonals are inclined toward the
center and resist only tension. ThePratt trussissuitablefor steel bridgessinceit istension that is
most effectively resisted. It should be noted, however, that vertical members of Pratt truss are in
compression. A Howetrussissimilar to thePratt except that thediagonalsareinclined toward the
ends, leadingto axial compression forces, and thevertical membersresist tension. Wooden bridges
often makeuseof theHowetrusssincetheconnectionsof thediagonalsin wood tend to compress.
AK-truss, so named sincethewebmembersforma K, ismost economical in largebridgesbecause
theshort member lengthsreducetherisk of buckling.
10.8.3 Structural AnalysisandSecondaryStress
Thetrussisaframed structureof bars, theoretically connected by hinges, forming stabletriangles.
Trusses contain triangle framed units to keep it stable. Its members are assumed to resist only
tensileor compressiveaxial forces. A statically determinatetrusscan beanalyzed usingequilibrium
conditionsonly. If morethan theleast number of membersrequired for stability areprovided, the
trussbecomesindeterminateand can no longer besolved usingonly theconditionsof equilibrium.
Thedisplacement compatibilityshouldbeadded. Aninternallyand/or externallyindeterminatetruss
isbest solved usingcomputer software.
Inpractice, trussmembersareconnectedtogusset plateswithhigh-tensionbolts(seeFigure10.42),
not rotation-freehinges, simplybecausethesearemucheasier tofabricate. Thepinned conditionof
theoryisnot reected in theeld. Thisdiscrepancyresultsin secondarystresses (bendingstresses)
in themembers. Secondary stressesaregiven by acomputer analysisof arigid frameand areusually
foundtobelessthan 20%of theprimary(axial) stresses. If thetrussmembersareproperlydesigned,
c 1999by CRCPressLLC
FIGURE10.41: Typesof trusses.
FIGURE10.42: Nodal jointsof atrussbridge. (FromJapanAssociationof Steel BridgeConstruction,
Outlineof Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
that is, theslendernessratiosof thetrussbarsaresufciently largewith no buckling, then secondary
stressescan conveniently and reliably bedisregarded.
10.8.4 Gerber TrussBridge
Figure10.43 isaphoto of aGerber trussbridgeduringtheerection of thecentral part, which isthe
Minato Oh-Hashi in Japan. Itsplan viewisshown in Figure10.2. A Gerber trusshasintermediate
hingesbetweenthesupportstocreateastaticallydeterminatestructural system. Inthecaseof Minato
Oh-Hashi, thesoil condition at thebottomof theharbor wasfound tobenot stiff and solid; thusthe
Gerber trussproved thewisest choice.
c 1999by CRCPressLLC
FIGURE 10.43: Lifting erection of the Minato Oh-Hashi, Japan. (Gerber bridge, 1974). (From
Hanshin Expressway PublicCorporation, TechnoGallery, Osaka, Japan, 1994. With permission.)
10.9 RigidFrameBridges(RahmenBridges)
10.9.1 Structural Features
Themembersarerigidly connected in rahmen structuresor rigid frames. Unlikethetrussand
thearch bridges, which will bediscussed in thefollowingsubsection, all themembersaresubjected
to both an axial forceand bendingmoments. Figure10.44showsvarioustypesof rahmen bridges.
Themembersof arigidframebridgearemuchlarger thanthoseinatypical building. Consequently,
stressconcentrationsoccur at thejunctionsof beamsand columnswhich must becarefully designed
usingniteelement analysesor experimental verication. Thesupportsof rahmen bridgesareeither
hingedor xed, makingit an externallyindeterminatestructure, andit isthereforenot suitablewhen
thefoundation islikelytosink. Thereactionsat supportsarehorizontal and vertical forcesat hinges,
with theaddition of abendingmoment at axed base.
10.9.2 Portal Frame
A portal frame is the simplest design (Figure 10.44a) and is widely used for the piers of elevated
highwaybridgesbecausethespaceunderneath can beeffectivelyusedfor other roadsor parkinglots.
Thesepierswereproved, in the1995KobeEarthquakein Japan, tobemoreresilient, that is, toretain
morestrength and absorb moreenergy than single-legged piers.
10.9.3 -Rahmen(StruttedBeamBridge)
The-rahmen design is usually used for bridges in mountainous regions wherethefoundation is
rm, passingover deep valleyswith arelatively longspan, or for bridgescrossingover expressways
(Figure10.44b). Asshown in thestructural layout of a-rahmen bridgein Figure10.45, thetwo
legssupport themain girders, inducingaxial compression in thecenter span of thegirder. Liveload
on thedeck istransmitted to themain girdersthrough theoor system. Intermediatehingesmay be
c 1999by CRCPressLLC
FIGURE10.44: Typesof rahmen bridges.
insertedinthegirderstomakeGerber girders. AV-legrahmenbridgeissimilar toa-rahmenbridge
but can span longer distanceswith no axial forcein thecenter span of thegirder (Figure10.44c).
10.9.4 Vierendeel Bridge
TheVierendeel bridgeisa rigid framewhoseupper and lower chordsareconnected rigidly to the
vertical members(Figure10.44d). All themembersaresubjected to axial and shear forcesaswell
asbending moments. Thisisinternally ahighly indeterminatesystem. Analysisof theVierendeel
framemust consider secondary stresses(seeSection 10.8.3). It ismorestiff than Langer or Lohse
arch bridgesin which somememberstakeonly axial forces.
10.10 ArchBridges
10.10.1 Structural Features
An arch rib actslikeacircular beamrestrained not only vertically but also horizontally at both ends,
and thusresultsin vertical and horizontal reactionsat thesupports. Thehorizontal reaction causes
axial compression in addition tobendingmomentsin thearch rib. Thebendingmomentscaused by
thehorizontal forcebalancesthosedueto gravity loads. Compared with theaxial force, theeffect of
c 1999by CRCPressLLC
FIGURE10.45: -rahmen bridge. (FromJapan Association of Steel BridgeConstruction, Outlineof
Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
thebendingmoment isusually small. That iswhy thearch isoften madeof materialsthat havehigh
compressivestrength, such asconcrete, stone, or brick.
10.10.2 Typesof Arches
An arch bridgeincludestheroad deck and thesupportingarch. Varioustypesof archesareshown in
Figure10.46. In thegure, thethick linerepresentsthememberscarryingbendingmoment, shear,
andaxial forces. Thethinlinerepresentsmemberstakingaxial forcesonly. Archbridgesareclassied
into the deck and the through-deck types according to the location of the road surface, as shown
in Figure10.46. Sincethedeck in both typesof bridgesissustained by either vertical columnsor
hangerstothearch, structurallythesameaxial forceaction, either compression or tension, isin effect
in themembers. Thedifferenceisthat thevertical membersof deck bridgestakecompressiveforces
and thehangersof through-deck bridgestaketension. Theliveload actson thearch only indirectly.
Abasicstructural typefor an arch isatwo-hingearch (seeFigure10.46a). Thetwo-hingearch has
onedegreeof indeterminacy externally becausetherearefour end reactions. If onehingeisadded at
thecrownof thearch, creatingathree-hingearch, it isrendereddeterminate. If theendsareclamped,
turningit into axed arch, it becomesindeterminateto thethird degree. Thetied arch issubtended
by two hingesby atieand simply supported (Figure10.46b). Thetied arch isexternally determinate
but internally has one degree of indeterminacy. The oor structures hang from the arch and are
isolated fromthetie. Other typesof arch bridgeswill bediscussed later in moredetail.
10.10.3 Structural Analysis
Almost all bridgedesign analyses, in thisageof super computingpower, useniteelement methods.
Theanalysisof an arch isbasically thesameasthat for aframe. Theweb membersareanalyzed as
truss bars which takeonly axial forces. Thearch rib and thegirdersareanalyzed aseither trusses
or beam-columns depending on thetypeof arch considered. Beam-columns takeaxial and shear
forcesand bendingmoments. An arch ribisusuallymadeup of straight piece-wisecomponents, not
curved segments, and it isso analyzed.
c 1999by CRCPressLLC
FIGURE10.46: Typesof arch bridges. (From Shimada, S., Journal of BridgeandFoundationEngi-
neering, 25(8), 1991[ in Japanese] . With permission.)
10.10.4 Langer Bridge
TheLanger archisanalyzedbyassumingthat thearchribtakesonlyaxial compression(Figure10.46c).
Thearch rib isthin, but thegirdersaredeep and resist moment and shear aswell asaxial tension.
Thegirders of theLanger bridgeareregarded asbeing strengthened by thearch rib. Figure10.47
showsthestructural componentsof aLanger bridge.
If diagonals areused in theweb, it is called a trussed Langer. Thedifferencebetween a trussed
Langer and astandard trussisthat thelower chord isagirder instead of just abar. TheLanger bridge
isalso determinateexternally and indeterminateinternally. Thedeck-typebridgeof theLanger is
often called areversed Langer.
10.10.5 LohseBridge
TheLohsebridgeisvery similar to theLanger bridgeexcept theLohsebridgecarriesitsresistanceto
bendingin thearch rib aswell asthegirder (Figure10.46d). By thisassumption, theLohsebridgeis
stiffer than theLanger. Thedistribution of bendingmomentsin thearch rib and thegirder depends
on thestiffnessratio of thetwo members, which isthedesignersdecision. TheLohsearch bridges
may bethought of astiered beams(seeFigure10.35) connected by vertical members. Thevertical
c 1999by CRCPressLLC
FIGURE10.47: Langer arch bridge. (From Japan Association of Steel BridgeConstruction, Outline
of Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
membersareassumed to takeonly axial forces. Aesthetically, theLohseismoreimposingthan the
Langer, and isthereforesuited to urban areaswhiletheLanger tsinto mountain areas.
10.10.6 TrussedArchandNielsenArchBridges
Generally diagonal members are not used in arch bridges, thus avoiding difculty in structural
analysis. However, recent advancementsin computer technology havechanged thisoutlook. New
typesof archbridges, suchasthetrussedarchin whichdiagonal trussbarsareusedinsteadof vertical
membersor theNielsen Lohsedesign in which tension rodsareused for diagonals, havenow been
introduced (seeFigure10.46e, f ). Diagonal web membersincreasethestiffnessof abridgemoreso
than vertical members.
All themembersof thetrussbridgetakeonly axial forces. On theother hand, thetrussed arch
bridge may resist bending in either the arch rib or the girder, or both. Since the diagonals of the
Nielsen Lohsebridgecarry only axial tension, they areprestressed by thedead load to compensate
for thecompression forcedueto theliveload.
10.11 Cable-StayedBridges
10.11.1 Structural Features
A cable-stayed bridgehangsthegirdersfrom diagonal cablesthat aretensioned from thetower, as
shown in Figure10.48. Thecablesof cable-stayed bridgesareanchored in thegirders. Thegirders
aremost often supported by movableor xed hinges. Dueto thediagonally tensioned cables, axial
forcesand bendingmomentsareimposed on thegirder and thetower. Thebendingmoment in the
girder isreduced when supported by thecables, and spanscan belonger than conventional girder
bridges(aslong as300 to 500 m). Themaximum span length isthe890 m of theTataraBridgein
Japan (seeTable10.1). Becauseof thewonder and beauty of thistype, itsdesign hasbeen copied
in even relatively small bridges including ones carrying only pedestrians. For long-span bridges,
stability under strong wind currents should be carefully considered in the design. The dynamic
effectsof wind and earthquakesmust bestudied analytically and experimentally. Wind tunnel tests
may benecessary to ensureexcessiveoscillation doesnot occur alongthelength of thebridgeor in
c 1999by CRCPressLLC
FIGURE10.48: Cable-stayed bridge. (FromJapan Association of Steel BridgeConstruction, Outline
of Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
thetower. Thecablesalso may resonatein thewind if they arethin and exible. In thiscase, devices
arenecessary to curb thevibration. Thestability of bridgesunder wind loadswill bediscussed in
moredetail in Section 10.12(seeFigure10.61).
10.11.2 Typesof Cable-StayedBridges
Cable-stayed bridgesmay beclassied by thehanging formation of thecableand theshapeof the
tower. Figure 10.49 illustrates three typical cable formations. Structurally, the radial cable most
effectivelydecreasestheaxial forcein thetower andgirders; however, difcultyin construction arises
dueto thestructural complexity at thetop of thetower. Thefan typeismorecommon becausethe
cableconnectionsat thetower aredistributed. Theharp typeisaesthetically themost pleasing.
FIGURE10.49: Typesof cableformation. (FromNagai, N., BridgeEngineering, Kyoritsu Publishing
Co., Tokyo, Japan [ in Japanese] , 1994. With permission.)
Figure10.50showsvarioustower designs. Asthespan length becomeslarge, columnssuch asthe
A, theH, or theupside-down Yshapeareselected; thesehavesignicant torsional resistance.
10.11.3 Structural Analysis
Thecable-stayed bridgeisusually analyzed usinglinear elastic frameanalysis. Thecableismodeled
asabar element with hinged ends. Figure10.51showstheowof gravity loads. Most of theload is
c 1999by CRCPressLLC
FIGURE 10.50: Types of towers. (From Nagai, N., BridgeEngineering, Kyoritsu Publishing Co.,
Tokyo, Japan [ in Japanese] , 1994. With permission.)
transmittedtothesubstructurethroughthecablesandthetower, but somegoestothegirder directly.
Thesmaller thebendingstiffnessof thegirder, thelesstheload istaken by thegirder. Asthetower
becomeshigher, thetension forceof thecablecan bereduced.
Becauseof thesag in thecabledueto itsown weight, areduced elastic modulusmay beused in
analysis. Thisreducedmodulusisslightlylower than theactual elasticmodulusof thecablematerial.
Thegirder and thetower aredesigned to takeaxial compression, bending, and shear. Sincethelarge
forcein thecableisconcentrated on thegirder and tower, stressconcentration at thoseconnections
should becarefully checked usingniteelement analysis. Takinginto theconsideration thefact that
thesupportsaresubjected to largenegativereactions(uplift), Pendel bearingsareused. These, as
mentionedpreviously(seeFigure10.25), arecomposedof an eye-bar andtwoendhinges, whichmay
movehorizontally and rotatefreely.
In thepreliminary design, thebridgeismodeled asaplaneframe. For thedetails, however, more
precise analyses such as three-dimensional stress analyses may be used. Nonlinear effects may be
taken into consideration for exiblelong-span bridges.
c 1999by CRCPressLLC
FIGURE10.51: Forceow in cable-stayed bridges. (From Nagai, N., BridgeEngineering, Kyoritsu
PublishingCo., Tokyo, Japan [ in Japanese] , 1994. With permission.)
10.11.4 TensioninCable
Oneof theimportant aspectsinthedesignof acable-stayedbridgeisthedeterminationof thetension
forceinthecable, whichisdirectlyrelatedtoforcesinthetower andthegirder. Control onthetension
forcein thecablesiscritical. Thepre-tension of thecablesmust beknown becauseit changesthe
stresses in the girder and the tower. Figure 10.52 shows the bending moment distribution under
dead loadsalongthebridgebeforeand after theprestressingforceisapplied. It can beseen that the
proper prestressreducesbendingmomentsin thegirder signicantly. If thevertical component of
thetension isselected to beequal to thereaction of thecontinuousgirder (supported at thejunction
of the cable and girder), the bending moment in the girder can be reduced to match that of the
continuousgirder.
FIGURE10.52: Bendingmoment distribution. (FromJapanSocietyof Civil Engineers, Cable-Stayed
BridgesTechnologyanditsChange, Tokyo, Japan [ in Japanese] , 1990. With permission.)
Thefollowingthreegeneral principlesareto beconsidered in determiningcabletension [19] :
1. Avoid having any bending moments (generated by dead loads) in the tower. This is
c 1999by CRCPressLLC
accomplishedbybalancingthehorizontal componentsof thecabletension in theleft and
right endsof thetower.
2. Keepthebendingmomentsinthegirder small. It dependsonthelocationandthedistance
between jointsto thecable. Small distances(such asamulti-cable) will result in small
bendingmomentsin thegirders.
3. Closethegirder byconnectingthecenter blocklastlywithout usinganycompellingforces.
Thecabletensionisselectedsuchthat zerosectional forceexistsat thecenter of thegirder.
10.12 SuspensionBridges
10.12.1 Structural Features
Suspension bridgesusetwo main cablessuspended between two towersand anchored to blocksat
theends. Figure10.53 showsthestructural componentsof asuspension bridge. Stiffening girders
FIGURE10.53: Suspension bridge. (From Japan Association of Steel BridgeConstruction, Outline
of Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
areeither trussor box type(seeFigure10.54) and hung from themain cablesusing hangers. The
suspensionbridgeismost suitablefor longspans. Table10.7isalist of theworldstenlongest bridges,
all of which aresuspension bridges. Thelongest istheAkashi Kaikyo Bridge, which hasamain span
of 1990.8m, in Japan. It wasoriginally designed with amain span of 1990m(Figure10.55), but was
extended by 0.8mwhen theKobeEarthquakecamecloseto thismark in 1995.
The ow of forces in a suspension bridge is shown in Figure 10.56. The load on the girder is
transmitted tothetowersthrough thehangersand themain cables, and then totheanchor blocks. It
c 1999by CRCPressLLC
FIGURE10.54: Typesof stiffening girders. (From Japan Association of Steel BridgeConstruction,
Outlineof Steel Bridges, Tokyo [ in Japanese] , 1985. With permission.)
TABLE10.7 TheWorlds10Longest Bridges
Center span
Rank Name (m) Country Year completed
1 Akashi Kaikyo Bridge 1990 Japan 1998(est.)
2 Great Belt East Bridge 1624 Denmark 1997(est.)
3 Humber Bridge 1410 England 1981
4 TsingMaBridge 1377 China 1997(est.)
5 Verrazano GateBridge 1298 U.S. 1964
6 Golden GateBridge 1280 U.S. 1937
7 MackinacStraitsBridge 1158 U.S. 1957
8 Minami Bisan-Seto Bridge 1100 Japan 1988
9 Faith Sulton Mehmet Bridge 1090 Turkey 1988
10 BosporusBridge 1074 Turkey 1973
FromHonshu Shikoku BridgeAuthority, Booklet andBrochures, Japan. With permission.
can beseen that anchor blocksareessential totakethehorizontal reaction forcefromthecables. The
gravity of theanchor blocksresiststheupward component of thecabletension force, and theshear
forcebetween theanchor blocksandthefoundation resiststhehorizontal component. Construction
difculty may arisewheresoil conditionsarepoor. Different from thecable-stayed bridge, no axial
forceisinduced in thegirdersof asuspension bridgeunlessit isaself-anchored suspension bridge
(seeFigure10.57d).
Thesagin themain cableaffectsthestructural behavior of thesuspension bridge: thesmaller the
sag, thelarger thestiffness of thebridgeand thereby largehorizontal forces areapplied to anchor
blocks. In general theratio of thesagto themain span isselected to beabout 1:10.
10.12.2 Typesof SuspensionBridges
Suspension bridgescan beclassied bythesupport condition of their stiffeninggirdersand themain
cable(Figure10.57). Thethree-span, two-hingetypeismost commonly used for highway bridges.
Thecontinuousgirder isoftenadoptedfor railroadbridgestoavoid knucklepoints, whichadversely
affect thetrains. When thesidespan isshort, thesingle-span typeisselected. Themain cablesof
self-anchoredbridgesarexedtothegirdersinsteadof totheanchor blocks, makingtheconstruction
of anchor blocksunnecessary; instead theaxial compression iscarried in thegirdersasin thecable-
stayed bridge. Therearespecial cases(such astheSevern Bridgein England) wherediagonal hangers
havebeen used.
10.12.3 Structural Analysis
If thedead load of thecableand thestiffeninggirdersisassumed to beuniformly distributed along
the bridge length, the deection of the cable is parabolic and all dead loads are supported by the
c 1999by CRCPressLLC
FIGURE10.55: Sideview of Akashi Kaikyo Bridge, Japan (1998 expected). (From Honshu Shikoku BridgeAuthority, Technology of Akashi Kaikyo
Bridge, Japan [ in Japanese] . With permission.)
c

1
9
9
9
b
y
C
R
C
P
r
e
s
s
L
L
C
FIGURE10.56: Forceow in a suspension bridge. (From Nagai, N., BridgeEngineering, Kyoritsu
PublishingCo., Tokyo, Japan [ in Japanese] , 1994. With permission.)
FIGURE10.57: Typesof suspension bridges. (FromNagai, N., BridgeEngineering, Kyoritsu Publish-
ingCo., Tokyo, Japan [ in Japanese] , 1994. With permission.)
cable. In thiscase, only liveloadsact on thegirder.
There are two analytical procedures: elastic theory, in which linear elastic material and small
displacement areassumed, and thedeection theory, which considersthedeection of thecabledue
to live loads. When the span becomes large, elastic theory is too conservative in its estimation of
bendingmoments.
10.12.4 CableDesign
For thecable, thehigh-strengthsteel wire, i.e., usually5mmin diameter withastrengthof 160to180
kg/mm
2
(1760N/mm
2
) andzinc-galvanized, isused. Thereareseveral typesof cables(Figure10.58):
strand rope, spiral rope, locked coil rope(LCR), and parallel wirestrand (PWS). ThePWSisused
most commonlyfor suspension bridges; thousandsof parallel wireelementsarebundledintoacircle
by asqueezingmachine, then wrapped with steel wireand painted.
Thewireistreatedbyanair spinning(AS) methodor theprefabricatedparallel wirestrandmethod
(PWSS). In the ASmethod, the 5-mm wire is elected by rounding between anchor blocks one by
oneuntil theprescribed number of wiresisobtained. In thePWSSmethod, astrand that bundles
c 1999by CRCPressLLC
FIGURE10.58: Typesof cables. (FromNagai, N., BridgeEngineering,KyoritsuPublishingCo., Tokyo,
Japan [ in Japanese] , 1994. With permission.)
100 to 200 wireelements is suspended between theanchor blocks by xing with a socket. In this
method, theconstruction period can beshort becausemorewiresareelected at onetimethan in the
ASmethod. Thethick strand ismorestableto wind but harder to handleduring construction. A
foothold (or catwalk) must beprovided under thecablefor theworkersto attach thecableband to
themain cable.
10.12.5 StiffeningGirder
Trussor boxtypegirdersareusedtostiffensuspensionbridges. Thegirder must becarefullydesigned
tohavesufcient stiffnessfor windstability. For verylongspanstrussesaremost effectiveinimproving
thestiffnessand stability (seeFigure10.54). Thebox girder isalso often adopted dueto itseaseof
fabrication.
10.12.6 Tower
Thetower isdesignedtobesubjectedtolargeaxial compression andbendingmoment. It isdesigned
to have smaller bending stiffness in the longitudinal direction since the horizontal forces coming
from both sidesof thetower keep it balanced. Figure10.59 showsa comparison of several towers
used for variousstructures. TheSearsTower in Chicago, known asthetallest building, hasaheight
of 443m.
A bridgetower usually consistsof morethan threecellsinside, each havingadequateresistanceto
torsion and local buckling under largeaxial forces. Mechanical damperssuch astheTMD (tuned
massdamper) or theTLD(tunedliquiddamper) areoften usedduringconstruction tocontrol tower
oscillationscaused by wind forces. Figure10.60showsatypical construction procedureadopted for
theAkashi Kaikyo Bridge, in which aclimbingtower craneisused. An alternativemethod isto use
acreeper crane, which clambersup alongthetower.
10.12.7 Stabilityfor Wind
Suspensionbridgesaresoexiblethat thedynamicstabilityunder windeffectsshouldbeinvestigated
usingawind tunnel. Thedynamicresponsesmay becategorized into threetypes, of which response
behaviorsareshown in Figure10.61: vortex-induced oscillations, buffeting, and torsional utters.
The utter, also called galloping, is torsional oscillation and is especially dangerous since it is a
self-diverging resonance and may incite failure quickly and easily. The ow of air increases the
amplitude of oscillations under certain combinations of wind speed and structural characteristics
(natural frequency), asillustratedinFigure10.61. Flexiblebridges, suchassuspensionor cable-stayed
bridges, must becarefully designed if thewind speedsarelikely to inciteutter.
c 1999by CRCPressLLC
FIGURE10.59: Comparison of towers. (From Honshu Shikoku Bridge Authority, Technology of
Akashi Kaikyo Bridge, Japan [ in Japanese] . With permission.)
Vortex-induced oscillations were once thought to be caused by the Karman vortex. Now it is
understood to betheair ow coming from thesurfaceor edgeof thegirdersthat yieldsvibrations
whichresonatewiththenatural frequencyof thestructure. Thisvibrationoccursat alowandrelatively
narrow rangeof wind speedsand doesnot develop dangerousdegreesof amplitudeamplication.
Buffetingisarandomvibrationcausedbyturbulenceintheair owor spontaneousgusts. Horizontal
movementsaredominant andtheamplitudesincreaseproportionallywith thesquareof windspeed.
10.13 DeningTerms
Abutment: An end support for abridgestructure.
Arch bridge: A bridgethat includestheroad deck and thesupportingarch.
Bridge: A structurethat crossesover ariver, bay, or other obstruction, permittingthesmooth
and safepassageof vehicles, trains, and pedestrians.
Cable-stayed bridge: Abridgeinwhichthesuperstructureishungfromthediagonal cablesthat
aretensioned fromthetower.
Cast-in-placeconcrete: Concrete placed in its nal position in the structure while still in a
plasticstate.
Compositegirder: A stell girder connected to a concrete deck so that they respond to force
effectsasaunit.
Deck (slab): A component, with or without wearingsurface, directly supportingwheel loads.
Floor system: A superstructurein which thedeck isintegral with itssupportingcomponents,
such asoor beamsand stringers.
Girder: Astructural component whoseprimary function isto resist loadsin exureand shear.
c 1999by CRCPressLLC
FIGURE10.60: Construction of atower. (From Honshu Shikoku BridgeAuthority, Technology of
Akashi Kaikyo Bridge, Japan [ in Japanese] . With permission.)
Generally, thistermisused for fabricated sections.
Girder bridge: A bridgesuperstructurethat consistsof aoor slab, girders, and bearings.
Inuenceline: A continuousor discretized function over asection of girder whosevalueat a
point, multipliedbyaloadactingnormal tothegirder at that point, yieldstheforceeffect
beingsought.
Lever rule: The static summation of moments about one point to calculate the reaction at a
second point.
LRFD (Load and ResistanceFactor Design): A method of proportioning structural compo-
nents (members, connectors, connecting elements, and assemblages) such that no ap-
plicable limit state is exceeded when the structure is subjected to all appropriate load
combinations.
c 1999by CRCPressLLC
FIGURE10.61: Dynamic responseof a tower against wind. (From Nagai, N., BridgeEngineering,
Kyoritsu PublishingCo., Tokyo, Japan [ in Japanese] , 1994. With permission.)
Precast member: Concreteelement cast in alocation other than itsnal position.
Prestressed concrete: Concretecomponentsin which thestressesand deformationsareintro-
duced by application of prestressingforces.
Rigid framebridge: Abridgeinwhichthesuperstructureandsubstructuremembersarerigidly
connected.
Segmental bridge: A bridgein which primary load-supportingmembersarecomposed of in-
dividual members called segments post-tensioned together to act as a monolithic unit
under loads.
Substructure: Structural partsof thebridgewhich providethehorizontal span.
Superstructure: Structural partsof thebridgewhich support thehorizontal span.
Suspension bridge: Abridgein which thesuperstructureissuspended by two main cablesand
anchored to end blocks.
Trussbridge: A bridgesuperstructurewhich consistsof aoor systemand main trusses.
Acknowledgment
Manyof theguresinthischapter arecopiedfromother booksandjournals. Theauthorswouldliketo
expresssinceregratitudeto theoriginal authors. Special thanksgo to Prof. N. Nagai of theNagaoka
Institute of Science and Technology, Profs. Y. Tachibana and H. Nakai of Osaka City University,
the Japan Association of Steel Bridge Construction, American Association of State Highway and
Transportation Ofcials, and CaliforniaDepartment of Transportation for their generosity.
References
[ 1] American Association of StateHighway and Transportation Ofcials. 1994. AASHTOLRFD
BridgeDesignSpecications, 1st ed., AASHTO, Washington, D.C.
[ 2] AmericanAssociationof StateHighwayandTransportationOfcials. 1989.GuideSpecications
for DesignandConstructionof Segmental ConcreteBridges, AASHTO, Washington, D.C.
[ 3] Caltrans. 1990. BridgeDesign Details Manual. California Department of Transportation,
Sacramento, CA.
c 1999by CRCPressLLC
[ 4] Caltrans.1993.BridgeDesignPracticeManual, vol.2, CaliforniaDepartment of Transportation,
Sacramento, CA.
[ 5] Caltrans. 1990. BridgeDesignAidsManual. CaliforniaDepartment of Transportation, Sacra-
mento, CA.
[ 6] Federal Highway Administration. 1990. StandardPlansfor HighwayBridges, Vol. I, Concrete
Superstructures, U.S. Department of Transportation, FHWA, Washington, D.C.
[ 7] Gerwick, B.C., Jr. 1993. Constructionof PrestressedConcreteStructures, 2nd ed., John Wiley
& Sons, NewYork.
[ 8] Hanshin Expressway Public Corporation. 1975. ConstructionRecordsof MinatoOh-Hashi,
HEPC, Japan Society of Civil Engineers, Tokyo, Japan (in Japanese).
[ 9] Hanshin Expressway PublicCorporation. 1994. TechnoGallery, HEPC, Osaka, Japan.
[ 10] Honshu Shikoku Bridge Authority. Technology of Akashi Kaikyo Bridge, HSBA, Japan (in
Japanese).
[ 11] Japan Association of Steel Bridge Construction. 1981. Manual Design Data Book, JASBC,
Tokyo, Japan (in Japanese).
[ 12] JapanAssociationof Steel BridgeConstruction. 1984. AGuideBookofBearingDesignforSteel
Bridges, JASBC, Tokyo, Japan (in Japanese).
[ 13] Japan Association of Steel BridgeConstruction. 1984. AGuideBookofExpansionJointDesign
for Steel Bridges, JASBC, Tokyo, Japan (in Japanese).
[ 14] Japan Association of Steel BridgeConstruction. 1985. Outlineof Steel Bridges, JASBC, Tokyo,
Japan (in Japanese).
[ 15] Japan Association of Steel BridgeConstruction. 1988. PlanningofSteel Bridges, JASBC, Tokyo,
Japan (in Japanese).
[ 16] Japan Construction Mechanization Association. 1991. Cost Estimation of BridgeErection,
JCMA, Tokyo, Japan (in Japanese).
[ 17] Japan RoadAssociation. 1993. Specicationsfor HighwayBridges, PartI CommonProvisions,
Part II Steel Bridges, andPart III ConcreteBridges, JRA, Tokyo, Japan (in Japanese).
[ 18] Japan Society of Civil Engineers. 1990. Cable-StayedBridgesTechnologyanditsChange,
JSCE, Tokyo, Japan (in Japanese).
[ 19] Nagai, N. 1994. BridgeEngineering, Kyoritsu PublishingCo., Tokyo, Japan (in Japanese).
[ 20] Podolny, W. andMuller, J.M. 1982. ConstructionandDesignofPrestressedConcreteSegmental
Bridges, John Wiley & Sons, NewYork.
[ 21] Shimada, S. 1991. Basictheoryof archstructures, Journal ofBridgeandFoundationEngineer-
ing, Kensetsu-Tosho, 25(8), 48-52, Tokyo, Japan (in Japanese).
[ 22] Tachibana, Y. and Nakai, H. 1996. BridgeEngineering, Kyoritsu PublishingCo., Tokyo, Japan
(in Japanese).
[ 23] Tonias, D.E. 1995. BridgeEngineering, McGraw-Hill, NewYork.
[ 24] Troitsky, M.S. 1994. PlanningandDesignof Bridges, John Wiley & Sons, NewYork.
[ 25] VSL. 1994. VSL Post-TensioningSystem, VSL Corporation, Campbell, CA.
[ 26] Xanthakos, P.P. 1994. TheoryandDesignof Bridges, John Wiley & Sons, NewYork.
[ 27] Xanthakos, P.P. 1995. BridgeSubstructureandFoundationDesign,Prentice-Hall, Upper Saddle
River, NJ.
Further Reading
[ 1] Billington, D.P. 1983. TheTower andtheBridge, BasicBooks, Inc., NewYork.
[ 2] Leonhardt, F. 1984. Bridges, AestheticsandDesign, MIT Press, Cambridge, MA.
[ 3] Chen, W. F. and Duan, L. 1998. Handbookof BridgeEngineering, CRCPress, BocaRaton, FL.
c 1999by CRCPressLLC
Appendix: DesignExamples
10.A.1Two-Span, Continuous, Cast-in-Place, PrestressedConcreteBoxGirder
Bridge
Given: A two-span, continuous, cast-in-place, prestressed concretebox girder bridgehastwo equal
spansof length 157 ft (47.9 m) with acolumn bent. Thesuperstructureis34 ft (10.4 m) wide. The
elevation of thebridgeisshown in Figure10.62a.
Material:
Initial concretef

ci
= 3500psi (24.13MPa), E
ci
= 3372 ksi (23,250MPa)
Final concretef

c
= 4000psi (27.58MPa), E
c
=3600ksi (24,860MPa)
Prestressing steel f
pu
= 270 ksi (1860 MPa) low relaxation strand, E
p
= 28,500 ksi
(197,000MPa)
Mild steel f
y
=60ksi (414MPa), E
s
=29,000ksi (200,000MPa)
Prestressing:
Anchorageset thickness=0.375in. (9.5mm)
Prestressingstressat jackingf
pj
=0.8
f
pu
=216ksi (1489MPa)
Thesecondary momentsdueto prestressing at thebent areM
DA
= 1.118 P
j
(kips-ft)
M
DG
=1.107P
j
(kips-ft)
Loads:
Dead load =self-weight +barrier rail +futurewearing3in ACoverlay
Liveload =AASHTOHS20-44+dynamicload allowance
Specication:
AASHTO-LRFD [ 1] (referred to asAASHTOin thisexample)
Requirements:
1. Determinecross-section geometry
2. Determinelongitudinal section and cablepath
3. Calculateloads
4. Calculateliveload distribution factors
5. Calculateunfactored momentsand shear demandsfor interior girder
6. Determineload factorsfor strength limit stateI and servicelimit stateI
7. Calculatesection propertiesfor interior girder
8. Calculateprestresslosses
9. Determineprestressingforce, P
j
, for interior girder
10. Check concretestrength for interior girder, servicelimit stateI
11. Flexural strength design for interior girder, strength limit stateI
12. Shear strength design for interior girder, strength limit stateI
c 1999by CRCPressLLC
FIGURE10.62: A two-span, continuous, prestressed concretebox girder bridge.
Solution
1. DetermineCross-Section Geometry
1.1) Structural Depth, d
For prestressed continuous spans, the structural depth, d, can be determined using a
c 1999by CRCPressLLC
depth-to-span ratio (d/L) of 0.04.
d = 0.04L = 0.04(157) = 6.28 ft (1.91 m)
Use d = 6.25 ft (1.91 m)
1.2) Girder Spacing, S
To provideeffectivetorsional resistanceand asufcient number of girdersfor prestress
paths, thespacingof girdersshould not belarger than twicetheir depth.
S
max
< 2d = 2(6.25) = 12.5 ft (3.81 m)
Using an overhang of 4 ft (1.22 m), thecenter-to-center distancebetween two exterior
girdersis26ft (7.92m).
Try threegirdersand two bays, S = 26/2 =13ft > 12.5ft N.G.
Try four girdersand threebays, S = 26/3 =8.67ft < 12.5ft O.K.
Useagirder spacing, S =8.67ft (2.64m)
1.3) Typical Section
From past experienceand design practice, weselect athicknessof 7in. (178mm) at the
edgeand 12 in. (305 mm) at thefaceof exterior girder for theoverhang, thewidth of
12 in. (305 mm) for girderswith theexterior girder aring to 18 in. (457 mm) at the
anchorageend. Thelengthof thisareisusuallytakenasone-tenthof thespanlength15.7
ft (4.79m). Thedeck andsoft thicknessesdependontheclear distancebetweenadjacent
girders. Wechoose7.875 in. (200 mm) and 5.875 in. (149 mm) for thedeck and soft
thicknesses, respectively. A typical section for this exampleis shown in Figure10.62b.
Thesection propertiesof thebox girder are:
Properties Midspan Bent (faceof support)
A ft
2
(m
2
) 57.25(5.32) 68.98(6.41)
I ft
4
(m
4
) 325.45(2.81) 403.56(3.48)
y
b
ft (m) 3.57(1.09) 3.09(0.94)
2. DetermineLongitudinal Section and CablePath
To lower the center of gravity of the superstructure at the face of a bent cap in a cast-
in-placepost-tensioned box girder, thethicknessof soft isared to 12 in., asshown in
Figure10.62c. Acablepath isgenerally controlled by themaximumdead load moments
and theposition of thejack at theend section. Maximum eccentricitiesshould occur at
pointsof maximum dead load moment and almost no eccentricity should bepresent at
thejacked end section. For thisexample, themaximumdead load momentsoccur at the
bent cap, closeto 0.4L for span 1 and 0.6L for span 2. A parabolic cablepath ischosen
asshown in Figure10.62c.
3. CalculateLoads
3.1) Component DeadLoad, DC
Thecomponent dead load, DC, includesall structural dead loadswith theexception of
thefuturewearingsurfaceand specied utility loads. For design purposes, two partsof
theDC aredened as:
DC1: girder self-weight (150lb/ft
3
) actingat theprestressingstate
DC2: barrier rail weight (784kips/ft) actingat servicestateafter all losses
c 1999by CRCPressLLC
3.2) WearingSurfaceLoad, DW
Thefuturewearingsurfaceof 3in. (76mm) with aunit weight of 140lb/ft
3
isdesigned
for thisbridge.
DW = (deck widthbarrier width) (thicknessof wearingsurface) (unit weight)
= [34 2(1.75)](0.25)(140) = 1067.5 lb/ft
3.3) LiveLoad, LL, andDynamicLoadAllowance, IM
The design live load, LL, is the AASHTO HS20-44 vehicular live load. To consider
thewheel load impact from moving vehicles, thedynamic load allowance, IM = 33%
(AASHTO-LRFD Table3.6.2.1-1), isused.
4. CalculateLiveLoad Distribution Factors
AASHTO-LRFD [ 1] recommends that approximatemethods beused to distributelive
load to individual girders. The dimensions relevant to this prestressed box girder are:
depth, d =6.25ft (1.91m); number of cells, N
c
=3; spacingof girders, S =8.67ft (2.64
m); span length, L = 157 ft (47.9 m); half of thegirder spacingplusthetotal overhang,
W
e
=8.334ft (2.54m); and thedistancebetween thecenter of an exterior girder and the
interior edgeof abarrier, d
e
= 41.75 = 2.25 ft (0.69 m). Thisbox girder iswithin the
rangeof applicability of theAASHTOapproximateformulas. Theliveload distribution
factorsarecalculated asfollows.
4.1) LiveLoadDistributionFactor for BendingMoments
(a) Interior girder (AASHTOTable4.6.2.2.2b-1)
Onelaneloaded:
LD
M
=
_
1.75 +
S
3.6
__
1
L
_
0.35
_
1
N
c
_
0.45
=
_
1.75 +
8.67
3.6
__
1
157
_
0.35
_
1
3
_
0.45
= 0.432 lanes
Two or morelanesloaded:
LD
M
=
_
13
N
c
_
0.3
_
S
5.8
__
1
L
_
0.25
=
_
13
3
_
0.3
_
8.67
5.8
__
1
157
_
0.25
= 0.656 lanes (controls)
(b) Exterior girder (AASHTOTable4.6.2.2.2d-1)
LD
M
=
W
e
14
=
8.334
14
= 0.595 lanes (controls)
4.2) LiveLoadDistributionFactor for Shear
(a) Interior girder (AASHTOTable4.62.2.3a-1)
c 1999by CRCPressLLC
Onelaneloaded:
LD
V
=
_
S
9.5
_
0.6
_
d
12L
_
0.1
=
_
8.67
9.5
_
0.6
_
6.25
12(157)
_
0.1
= 0.535 lanes
Two or morelanesloaded:
LD
V
=
_
S
7.3
_
0.9
_
d
12L
_
0.1
=
_
8.67
7.3
_
0.9
_
6.25
12(157)
_
0.1
= 0.660 lanes (controls)
(b) Exterior girder (AASHTOTable4.62.2.3b-1)
Onelaneloaded: Lever rule
Thelever ruleassumesthat thedeck in itstransversedirection issimply supported
by thegirdersand usesstaticsto determinetheliveload distribution to thegirders.
AASHTO-LRFDalsorequiresthat whenthelever ruleisused, themultiplepresence
factor, m, should apply. For aoneloaded lane, m = 1.2. Thelever rulemodel for
theexterior girder isshown in Figure10.63. Fromstaticequilibrium:
R =
5.92
8.67
= 0.683
LD

= mR = 1.2(0.683) = 0.820 (controls)

FIGURE10.63: Liveload distribution for exterior girderlever rule.
c 1999by CRCPressLLC
Two or morelanesloaded: Modify interior girder factor by e
LD
V
= e(LD
v
)
interior girder
=
_
0.64 +
d
e
12.5
_
(LD

)
interior girder
=
_
0.64 +
2.25
12.5
_
(0.66) = 0.541
Theliveload distribution factorsat thestrength limit state:
Strength limit stateI Interior girder Exterior girder
Bendingmoment 0.656lanes 0.595lanes
Shear 0.660lanes 0.820lanes
5. CalculateUnfactored Momentsand Shear Demandsfor Interior Girder
It is practically assumed that all dead loads are carried by the box girder and equally
distributed to each girder. Theliveloadstakeforcesto thegirdersaccordingto liveload
distributionfactors(AASHTOArticle4.6.2.2.2). Unfactoredmoment andshear demands
for aninterior girder areshowninFigures10.64and10.65. DetailsarelistedinTables10.8
and10.9. Onlytheresultsfor span 1areshown in thesetablesandguressincethebridge
issymmetrical about thebent.
TABLE10.8 Moment and Shear Dueto Unfactored Dead
Load for theInterior Girder (Span 1)
Unfactored dead load
DC1
a
DC2
b
DW
c
Location M
DC1
V
DC1
M
DC2
V
DC2
M
DW
V
DW
(x/L) (k-ft) (kips) (k-ft) (kips) (k-ft) (kips)
0.0 0 125.2 0 11.4 0 15.6
0.1 1700 91.5 155 8.4 212 11.4
0.2 2871 57.7 262 5.3 357 7.2
0.3 3513 24.0 321 2.2 437 3.0
0.4 3626 9.7 331 0.9 451 1.2
0.5 3210 43.4 293 4.0 399 5.4
0.6 2264 77.1 207 7.1 282 9.6
0.7 789 111 72 10.1 98 13.8
0.8 1215 145 111 13.2 151 18.0
0.9 3748 178 342 16.3 466 22.2
1.0 6833 216 622 19.4 847 26.4
(6292) (573) (781)
Note: Momentsin bracketsarefor faceof support at thebent. Momentsin
span 2 aresymmetrical about thebent. Shear in span areantisym-
metrical about thebent.
a
DC1, interior girder self-weight.
b
DC2, barrier self-weight.
c
DW, wearingsurfaceload.
6. DetermineLoad Factorsfor Strength Limit StateI and ServiceLimit StateI
6.1) General DesignEquation(AASHTOArticle1.3.2)

i
Q
i
R
n
(10.18)
c 1999by CRCPressLLC
FIGURE10.64: Moment envelopesfor span 1.
FIGURE10.65: Shear envelopesfor span 1.
c 1999by CRCPressLLC
TABLE10.9 Moment and Shear Envelopesand Associated Forcesfor the
Interior Girder Dueto AASHTOHS20-44LiveLoad (Span 1)
Positivemoment
and associated Negativemoment Shear and
shear and associated shear associated moment
Location M
LL+IM
V
LL+IM
M
LL+IM
V
LL+IM
V
LL+IM
M
LL+IM
(x/L) (k-ft) (kips) (k-ft) (kips) (kips) (k-ft)
0.0 0 0 0 0 60.0 0
0.1 782 49.8 85 5.4 50.1 787
0.2 1312 41.8 169 5.4 42.0 1320
0.3 1612 29.3 253 5.4 34.3 1614
0.4 1715 21.8 337 5.4 27.7 1650
0.5 1650 30.0 422 5.4 35.1 1628
0.6 1431 36.7 506 5.4 42.0 1424
0.7 1081 42.6 590 5.4 49.9 852
0.8 647 47.8 748 8.3 59.2 216
0.9 196 32.9 1339 50.1 68.8 667
1.0 0 0 2266 67.8 78.5 1788
(2104)
Note: LL + IM = AASHTO HS20-44 live load plus dynamic load allowance. Moments in
bracketsarefor faceof support at thebent. Momentsin span 2 aresymmetrical about
thebent. Shear inspan2isantisymmetrical about thebent. Liveloaddistributionfactors
areconsidered.
where
i
areload factors, is a resistancefactor, Q
i
represents forceeffects, R
n
is the
nominal resistance, and is a factor related to ductility, redundancy, and operational
importanceof that beingdesigned. isdened as:
=
D

I
0.95 (10.19)
where

D
=
_
1.05 for nonductilecomponentsand connections
0.95 for ductilecomponentsand connections
(10.20)

R
=
_
1.05 for nonredundant members
0.95 for redundant members
(10.21)

I
=
_

_
1.05 operationally important bridge
0.95 general bridge
only apply to strength and extreme
event limit states
(10.22)
For thisbridge, thefollowingvaluesareassumed:
Ductility Redundancy Importance
Limit states
D

R

I

Strengthlimit state 0.95 0.95 1.05 0.95
Servicelimit state 1.0 1.0 1.0 1.0
6.2) LoadFactorsandLoadCombinations
Theload factorsand combinationsarespecied as(AASHTOTable3.4.1-1):
Strength limit stateI: 1.25(DC1 +DC2) +1.5(DW) +1.75(LL +M)
Servicelimit stateI: DC1 +DC2 +DW +(LL +IM)
c 1999by CRCPressLLC
7. CalculateSection Propertiesfor Interior Girder
For an interior girder asshown in Figure10.66, theeffectiveangewidth, b
eff
, isdeter-
mined (AASHTOArticle4.6.2.6) by
b
eff
= thelesser of
_
_
_
L
eff
4
12h
f
+b
w
S
(10.23)
whereL
eff
is theeffectivespan length and may betaken as theactual span length for
simply supported spans and the distance between points of permanent load inection
for continuousspans; h
f
isthecompression angedepth; and b
w
istheweb width. The
effectiveangewidth and thesection propertiesareshown in Table10.10for theinterior
girder.
FIGURE10.66: Effectiveangewidth of interior girder.
TABLE10.10 EffectiveFlangeWidth and Section Propertiesfor
Interior Girder
Bent
Location Dimension Midspan (faceof support)
h
f
in. (mm) 7.875(200) 7.875(200)
L
eff
/4 in. (mm) 353(8966) 235.5(11963)
Top 12h
f
+b
w
in. (mm) 106.5(2705) 106.5(2705)
ange S in. (mm) 104(2642) 104(2642)
b
eff
in. (mm) 104(2642) 104(2642)
h
f
in. (mm) 5.875(149) 12(305)
L
eff
/4 in. (mm) 353(8966) 235.5(11963)
Bottom 12h
f
+b
w
in. (mm) 82.5(2096) 156(2096)
ange S in. (mm) 104(2642) 104(2642)
b
eff
in. (mm) 82.5(2096) 104(2096)
Area A ft
2
(m
2
) 14.38(1.336) 19.17(1.781)
Moment of inertia I ft
4
(m
4
) 81.85(0.706) 112.21(0.968)
C.G. y
b
ft (m) 3.55(1.082) 2.82(0.860)
Note:
L
eff
= 117.8ft (35.9m) for midspan,
L
eff
= 78.5ft (23.9m) for thebent,
b
w
= 12in. (305mm).
c 1999by CRCPressLLC
8. CalculatePrestressLosses
For a cast-in-placepost-tensioned box girder, two types of losses, instantaneous losses
(friction, anchorage set, and elastic shortening) and time-dependent losses (creep and
shrinkageof concreteand relaxation of prestressingsteel) aresignicant. Sincethepre-
stresslossesarenot symmetrical about thebent for thisbridge, thecalculationisperformed
for both spans.
8.1) Frictional Loss, f
pF
f
pF
= f
pj
_
1 e
(Kx+)
_
(10.24)
where K is the wobble friction coefcient = 0.0002 1/ft (6.6 10
7
1/mm); is the
coefcient of friction =0.25(AASHTOArticle5.9.5.2.2a); x isthelengthof aprestressing
tendon from thejacking end to thepoint considered; and isthesum of theabsolute
valuesof anglechangein theprestressingsteel path fromthejackingend.
For aparaboliccablepath(Figure10.67), theanglechangeis = 2e
p
/L
p
, wheree
p
isthe
vertical distancebetween two control pointsand L
p
isthehorizontal distancebetween
two control points. Thedetailsaregiven in Table10.11.
FIGURE10.67: Paraboliccablepath.
TABLE10.11 PrestressFrictional Loss
Segment e
p
(in.) L
p
(ft) (rad)

(rad)

L
p
(ft) Point f
pF
(ksi)
A 31.84 0 0 0 0 A 0.0
AB 31.84 62.8 0.0845 0.0845 62.8 B 7.13
BC 42.50 78.5 0.0902 0.1747 141.3 C 14.90
CD 8.50 15.7 0.0902 0.2649 157.0 D 20.09
DE 8.50 15.7 0.0902 0.3551 172.7 E 25.06
EF 42.50 78.5 0.0902 0.4453 251.2 F 32.18
FG 31.84 62.8 0.0845 0.5298 314.0 G 38.23
8.2) AnchorageSet Loss, f
pA
The effect of anchorage set on the cable stress can be approximated by the Caltrans
c 1999by CRCPressLLC
procedure[ 4] , asshown in Figure10.68. It isassumedthat theanchorageset losschanges
linearly within thelength, L
pA
.
FIGURE10.68: Anchorageset lossmodel. (From CaliforniaDepartment of Transportation, Bridge
DesignPractice, Copyright 1983 (Figure310, pages346, updated March, 1993), Sacramento, CA,
1993. With permission.)
L
pA
=
_
E(L)L
pF
f
pF
(10.25)
f =
2f
pF
x
L
pF
(10.26)
f
pA
= f
_
1
x
L
pA
_
(10.27)
where L is the thickness of the anchorage set; E is the modulus of elasticity of the
anchorageset; f isthechangeinstressduetotheanchor set; L
pA
isthelengthinuenced
bytheanchor set; L
pF
isthelengthtoapoint wherelossisknown; andx isthehorizontal
distancefromthejackingend to thepoint considered.
For an anchor set thicknessof L = 0.375 in. and E =29,000ksi, consider thepoint B
whereL
pF
= 141.3 ft and f
pF
= 14.9 ksi:
L
pA
=
_
E(L)L
pF
f
pF
=
_
29,000(3/8)(141.3)
12(14.90)
= 92.71 ft < 141.3 ft O.K
f =
2f
pF
x
L
pF
=
2(14.90)(92.71)
141.3
= 19.55 ksi
f
pA
= f
_
1
x
L
pA
_
= 19.55
_
1
x
92.71
_
8.3) ElasticShorteningLoss, f
pES
The loss due to elastic shortening in post-tensioned members is calculated using the
followingformula(AASHTOArticle5.9.5.2.3b):
c 1999by CRCPressLLC
f
pES
=
N 1
2N
E
p
E
ci
f
cgp
(10.28)
whereN isthenumber of identical prestressingtendonsandf
cgp
isthesumof theconcrete
stressat thecenter of gravityof theprestressingtendonsduetotheprestressingforceafter
jacking and theself-weight of member at thesection with themaximum moment. For
post-tensioned structures with bonded tendons, f
cgp
may be calculated at the center
section of thespan for simplysupportedstructuresandat thesection with themaximum
moment for continuousstructures. To calculatetheelastic shortening loss, weassume
that theprestressingjack forcefor an interior girder P
j
=1800kipsandthetotal number
of prestressingtendonsN =4. f
cgp
iscalculated for themid-support section:
f
cgp
=
p
j
A
+
P
j
e
2
I
x
+
M
DC1
e
I
x
=
1800
19.17(12)
2
+
1800(28.164)
2
112.21(12)
4
+
(6292)(12)(28.164)
112.21(12)
4
= 0.652 +0.614 0.914 = 0.352 ksi (2448 MPa)
f
pES
=
N 1
2N
E
p
E
ci
f
cgp
=
4 1
2(4)
28,500
3370
(0.352) = 1.12 ksi (7.7 MPa)
8.4) Time-Dependent Losses, f
pT M
AASHTO providesa tableto estimatetheaccumulated effect of time-dependent losses
resultingfromthecreepandshrinkageof concreteandtherelaxation of thesteel tendons.
FromAASHTOTable5.9.5.3-1:
f
pT M
= 21 ksi (145 MPa) (upper bound)
8.5) Total Losses, f
pT
f
pT
= f
pF
+f
pA
+f
pES
+f
pT M
Detailsaregiven in Table10.12.
9. DeterminePrestressingForce, P
j
, for Interior Girder
Since the live load is not in general equally distributed to the girders, the prestressing
force, P
j
, required for each girder may differ. To calculate prestress jacking force, P
j
,
theinitial prestressforcecoefcient, F
pCI
, andnal prestressforcecoefcient, F
pCF
, are
dened as:
F
pCI
= 1
f
pF
+f
pA
+f
pES
f
pj
(10.29)
F
pCF
= 1
f
pT
f
pj
(10.30)
Thesecondary moment coefcientsaredened as:
M
sC
=
_
x
L
M
DA
P
j
for span 1
(1
x
L
)
M
DG
P
j
for span 2
(10.31)
c 1999by CRCPressLLC
TABLE10.12 CablePath and PrestressLosses
Location Prestresslosses(ksi) Forcecoeff.
Span (x/L) f
pF
f
pA
f
pES
f
pT M
f
pT
F
pCI
F
pCF
0.0 0.00 19.55 41.67 0.904 0.807
0.1 1.78 16.24 40.14 0.911 0.814
0.2 3.56 12.93 38.61 0.918 0.821
0.3 5.35 9.93 37.40 0.924 0.827
0.4 7.13 6.31 35.56 0.933 0.835
1 0.5 8.68 3.00 1.12 21 33.79 0.941 0.844
0.6 10.24 32.36 0.947 0.850
0.7 11.79 33.91 0.940 0.843
0.8 13.35 0.00 35.47 0.933 0.836
0.9 14.90 37.02 0.926 0.829
1.0 20.09 42.21 0.902 0.805
0.0 20.09 42.21 0.902 0.805
0.1 25.06 47.18 0.879 0.782
0.2 26.49 48.61 0.872 0.775
0.3 27.91 50.03 0.866 0.768
0.4 29.34 51.46 0.859 0.762
0.5 30.76 0.00 1.12 21 52.88 0.852 0.755
2 0.6 32.18 54.30 0.846 0.749
0.7 33.69 55.81 0.839 0.742
0.8 35.21 57.33 0.832 0.735
0.9 36.72 58.84 0.825 0.728
1.0 38.23 60.35 0.818 0.721
Note:
F
pCI
= 1
f
pF
+f
pA
+f
pES
f
pj
F
pCF
= 1
f
pT
f
pj
where x is the distance from the left end for each span. The combined prestressing
moment coefcientsaredened as:
M
psCI
= F
pCI
(e) +M
sC
(10.32)
M
psCF
= F
pCF
(e) +M
sC
(10.33)
where e is the distance between the cable and the center of gravity of a cross-section;
positivevaluesof e indicatethat thecableisabovethecenter of gravity, andnegativeones
indicatethecableisbelowit.
Theprestressforcecoefcientsandthecombinedmoment coefcientsarecalculatedand
listedinTable10.13. AccordingtoAASHTO, theprestressingforce, P
j
, canbedetermined
usingtheconcretetensilestresslimit in theprecompression tensilezone(seeTable10.5):
f
DC1
+f
DC2
+f
DW
+f
LL+IM
+f
psF
0.19
_
f

c
(10.34)
in which
c 1999by CRCPressLLC
TABLE10.13 PrestressForceand Moment Coefcients
Location Cablepath Forcecoeff. Moment coefcients(ft)
Span (x/L) e (in.) F
pCI
F
pCF
F
pCI
e F
pCF
e M
sC
M
psCI
M
psCF
0.0 0.240 0.904 0.807 0.018 0.016 0.000 0.018 0.016
0.1 13.692 0.911 0.814 1.040 0.929 0.112 0.928 0.817
0.2 23.640 0.918 0.821 1.809 1.618 0.224 1.586 1.394
0.3 29.136 0.924 0.827 2.244 2.008 0.335 1.908 1.672
0.4 31.596 0.933 0.835 2.456 2.200 0.447 2.008 1.752
1 0.5 29.892 0.941 0.844 2.344 2.101 0.559 1.785 1.542
0.6 24.804 0.947 0.850 1.958 1.757 0.671 1.287 1.087
0.7 13.608 0.940 0.843 1.278 1.146 0.783 0.495 0.363
0.8 4.404 0.933 0.836 0.342 0.307 0.894 0.552 0.588
0.9 10.884 0.926 0.829 0.840 0.752 1.006 1.846 1.758
1.0 28.164 0.902 0.805 2.117 1.888 1.118 3.235 3.006
0.0 28.164 0.902 0.805 2.117 1.888 1.107 3.224 2.995
0.1 10.884 0.879 0.782 0.797 0.709 0.996 1.793 1.705
0.2 4.404 0.872 0.775 0.320 0.284 0.886 0.566 0.601
0.3 16.308 0.866 0.768 1.176 1.044 0.775 0.401 0.269
0.4 24.804 0.859 0.762 1.776 1.575 0.664 1.111 0.910
0.5 29.892 0.852 0.755 2.123 1.881 0.554 1.570 1.328
2 0.6 31.596 0.846 0.749 2.227 1.971 0.443 1.784 1.528
0.7 29.136 0.839 0.742 2.037 1.801 0.332 1.705 1.469
0.8 23.640 0.832 0.735 1.639 1.447 0.221 1.417 1.226
0.9 13.692 0.825 0.728 0.941 0.830 0.111 0.830 0.719
1.0 0.240 0.818 0.721 0.016 0.014 0.000 0.016 0.014
Note: e is the distance between the cable path and central gravity of the interior girder cross-section; positive
meanscableisabovethecentral gravity and negativeindicatescableisbelowthecentral gravity.
f
DC1
=
M
DC1
C
I
x
(10.35)
f
DC2
=
M
DC2
C
I
x
(10.36)
f
DW
=
M
DW
C
I
x
(10.37)
f
LL+IM
=
M
LL+IM
C
I
x
(10.38)
f
psF
=
P
pe
A
+
(P
pe
e)C
I
x
+
M
s
C
I
x
=
F
pCF
P
j
A
+
M
psCF
P
j
C
I
x
(10.39)
whereC(= y
b
or y
t
) isthedistancefromtheextremeber to thecenter of gravity of the
cross-section; f

c
isin ksi; and P
pe
istheeffectiveprestressing forceafter all losseshave
been incurred. FromEquations10.34and 10.39, wehave:
P
j
=
f
DC1
f
DC2
f
DW
f
LL+IM
0.19
_
f

c
F
pCF
A
+
M
psCF
C
I
x
(10.40)
DetailedcalculationsaregiveninTable10.14. Most critical pointscoincidewithlocations
of maximumeccentricity: 0.4Linspan1, 0.6Linspan2, andat thebent. For thisbridge,
thecontrollingsectionisthroughtheright faceof thebent. Herein, P
j
=1823kips(8109
kN). Rounding P
j
up to 1830 kips(8140 kN) givesarequired areaof prestressing steel
of A
ps
= P
j
/f
pj
= 1830/216 =8.47in.
2
(5465mm
2
).
10. Check ConcreteStrength for Interior Girder, ServiceLimit StateI
Two criteria are imposed on the level of concrete stresses when calculating required
concretestrength (AASHTOArticle5.9.4.2):
c 1999by CRCPressLLC
TABLE10.14 Determination of PrestressingJackingForcefor an Interior Girder
Top ber Bottomber
Jacking Jacking
Location Stress(psi) force Stress(psi) force
Span (x/L) f
DC1
f
DC2
f
DW
f
LL+IM
P
j
(kips) f
DC1
f
DC2
f
DW
f
LL+IM
P
j
(kips)
0.0 0 0 0 0 0 0 0 0
0.1 389 36 48 179 512 47 64 236 749
0.2 658 60 82 301 865 79 108 395 1307
0.3 805 73 100 369 1058 97 132 485 1542
0.4 831 76 103 393 1092 100 136 517 1573
1 0.5 735 67 91 378 967 88 120 497 1482
0.6 519 47 64 328 682 62 85 431 1193
0.7 181 16 22 248 238 22 30 326 455
0.8 278 25 35 171 242 366 33 46 225
0.9 859 78 107 307 1210 1129 103 140 403
1.0 1336 122 166 447 1818 1098 100 136 367
0.0 1336 122 166 447 1823 1098 100 136 367
0.1 859 78 107 307 1264 1129 103 140 403
0.2 278 25 35 171 254 366 33 46 225
0.3 181 16 22 248 238 22 30 326 520
0.4 519 47 64 328 682 62 85 431 1371
2 0.5 735 67 91 378 967 88 120 497 1691
0.6 831 76 103 393 1092 100 136 517 1782
0.7 805 73 100 369 1058 97 132 485 1739
0.8 658 60 82 301 865 79 108 395 1474
0.9 389 36 48 179 512 47 64 236 843
1.0 0 0 0 0 0 0 0 0
Note: Positivestressindicatescompression and negativestressindicatestension. P
j
areobtained by Equation 10.40.
_
f
DC1
+f
psI
0.55f

ci
at prestressingstate
f
DC1
+f
DC2
+f
DW
+f
LL+IM
+f
psF
0.45f

c
at servicestate
(10.41)
f
psI
=
P
jI
A
+
_
P
jI
e
_
C
I
x
+
M
sI
C
I
x
=
F
pCI
P
j
A
+
M
psCI
P
j
C
I
x
(10.42)
Theconcretestressesin theextremebers(after instantaneouslossesandnal losses) are
given in Tables10.15and 10.16. For theinitial concretestrength in theprestressingstate,
thecontrollinglocationisthebottomber at 0.9Lsectioninspan1. FromEquation10.41
wehave:
f

ci,req

f
DC1
+f
psI
0.55
=
930
0.55
= 1691 psi < 3500 psi
.
.. choose f

ci
= 3500 psi (24.13 MPa) O.K.
For thenal concretestrength at theservicelimit state, thecontrollinglocation isagain
in thebottomber at 0.9L section in span 1. FromEquation 10.41wehave:
f

c,req

f
DC1
+f
DC2
+f
DW
+f
LL+IM
+f
psF
0.45
=
1539
0.45
= 3420 psi < 4000 psi
.
.. choose f

c
= 4000 psi (27.58 MPa) O.K.
11. Flexural Strength Design for Interior Girder, Strength Limit StateI
AASHTOrequiresthat for thestrength limit stateI
c 1999by CRCPressLLC
TABLE10.15 ConcreteStressesafter InstantaneousLossesfor theInterior Girder
Top ber stress(psi) Bottomber stress(psi)
Total Total
Location F

pCI
M

psCI
initial F

pCI
M

psCI
initial
Span (x/L) f
DC1
P
j
/A P

j
Y
t
/I f
psI
stress f
DC1
P
j
/A P

j
y
h
/I f
psI
stress
0.0 0 799 8 807 807 0 799 10 789 789
0.1 389 805 389 416 806 512 805 512 1317 805
0.2 658 812 665 147 805 865 812 874 1686 821
0.3 805 817 800 17 821 1058 817 1052 1868 810
0.4 831 824 842 18 813 1092 824 1107 1931 839
1 0.5 735 831 748 83 819 967 831 984 1815 848
0.6 519 837 540 298 816 682 837 710 1547 865
0.7 181 831 208 623 804 238 831 723 1104 866
0.8 278 825 231 1056 778 366 825 304 520 866
0.9 859 818 774 1592 733 1129 818 1017 199 930
1.0 1336 598 1257 1854 519 1098 598 1033 435 663
0.0 1336 598 1252 1850 514 1098 598 1030 432 666
0.1 859 777 752 1528 670 1129 777 988 212 917
0.2 278 771 237 1008 730 366 771 312 459 825
0.3 181 765 168 597 777 238 765 221 986 749
0.4 519 759 466 293 812 682 759 613 1372 690
2 0.5 735 753 658 95 830 967 753 865 1619 652
0.6 831 748 748 0 830 1092 748 983 1731 639
0.7 805 741 715 27 831 1058 741 940 1681 623
0.8 658 735 594 141 799 865 735 781 1516 651
0.9 389 729 348 381 770 512 729 458 1187 675
1.0 0 723 7 730 730 0 723 9 714 714
Note: Positivestressindicatescompression and negativestressindicatestension.
TABLE10.16 ConcreteStressesafter Total Lossesfor theInterior Girder
Top ber stress(psi) Bottomber stress(psi)
Total Total
Location F

pCF
M

psCF
nal F

pCF
M

psCF
nal
Span (x/L) f
LOAD
P
j
/A P

j
Y
t
/I f
psF
stress f
LOAD
P
j
/A P

j
y
b
/I f
psF
stress
0.0 0 713 7 720 720 0 713 9 704 704
0.1 653 720 343 377 1030 858 720 450 1170 312
0.2 1100 726 584 141 1241 1466 726 768 1494 48
0.3 1348 731 701 30 1377 1772 731 922 1652 119
0.4 1403 738 735 4 1407 1844 738 966 1704 140
1 0.5 1272 746 647 99 1371 1672 746 850 1596 76
0.6 958 751 455 296 1254 1260 751 599 1350 90
0.7 467 745 152 593 1060 614 745 200 945 331
0.8 510 739 246 985 475 670 739 324 415 1085
0.9 1351 732 737 1469 119 1776 732 969 237 1539
1.0 2070 533 1168 1701 368 1702 533 960 427 1275
0.0 2070 533 1164 1697 373 1702 533 957 423 1278
0.1 1351 691 715 1406 55 1776 691 940 249 1527
0.2 510 685 252 937 427 670 685 331 353 1024
0.3 467 679 113 566 1033 614 679 148 828 213
0.4 958 673 382 292 1250 1260 673 502 1175 85
2 0.5 1272 667 557 111 1383 1672 667 732 1399 273
0.6 1403 662 641 21 1424 1844 662 842 1504 340
0.7 1348 655 616 40 1387 1772 655 809 1465 307
0.8 1100 649 514 135 1235 1466 649 676 1325 122
0.9 653 643 302 341 994 858 643 397 1040 181
1.0 0 637 6 643 643 0 637 8 629 629
Note: f
LOAD
= f
DC1
+f
DC2
+F
DW
+f
LL+IM
. Positivestressindicatescompression and negativestressindicatestension.
M
u
M
n
M
u
=

i
M
i
= 0.95 [1.25(M
DC1
+M
DC2
)
+ 1.5M
DW
+1.75M
LLH
] +M
ps
where is the exural resistance factor 1.0 and M
ps
is the secondary moment due to
c 1999by CRCPressLLC
prestress. Factoredmoment demands, M
u
, for theinterior girder in span 1arecalculated
in Table 10.17. Although the moment demands are not symmetrical about the bent
(due to different secondary prestress moments), the results for span 2 are similar and
thedifferenceswill not beconsidered in thisexample. Thedetailed calculationsfor the
exural resistance, M
n
, areshown in Table10.18. It isclear that noadditional mildsteel
isrequired.
TABLE10.17 Factored Momentsfor an Interior Girder (Span 1)
M
DC1
M
DC2
M
DW
(kips-ft) (kips-ft) (kips-ft) M
LL+IM
M
ps
M
u
Location Dead Dead Wearing (kips-ft) (kips-ft) (kips-ft)
(x/L) load-1 load-2 surface Positive Negative P/S Positive Negative
0.0 0 0 0 0 0 0 0 0
0.1 1700 155 212 782 85 205 4009 2569
0.2 2871 262 357 1312 169 409 6820 4358
0.3 3513 321 437 1612 253 614 8469 5368
0.4 3626 331 451 1715 337 818 9012 5599
0.5 3210 293 399 1650 422 1023 8494 5050
0.6 2264 207 282 1431 506 1228 6942 3721
0.7 789 72 98 1081 590 1432 4392 1613
0.8 1215 111 151 647 748 1637 922 1397
0.9 3748 342 466 196 1339 1841 3355 5906
1.0 6292 573 781 0 2104 2046 7219 10716
Note: M
u
= 0.95[1.25(M
DC1
+M
MDC2
) +1.5M
DW
+1.75M
LL+IM
] +M
ps
TABLE10.18 Flexural Strength Design for Interior Girder, Strength Limit StateI (Span 1)
Location A
ps
d
p
A
s
d
s
b c f
ps
d
e
a M
n
M
u
(x/L) (in.
2
) (in.) (in.
2
) (in.) (in.) (in.) (ksi) (in.) (in.) (k-ft) (k-ft)
0.0 32.16 0 72.06 104 7.14 253.2 32.16 6.07 5206 0
0.1 46.09 0 72.06 104 7.27 258.1 46.09 6.18 7833 4009
0.2 56.04 0 72.06 104 7.33 260.1 56.04 6.23 9717 6820
0.3 61.54 0 72.06 104 7.35 261.0 61.54 6.25 10759 8469
0.4 64.00 0 72.06 104 7.36 261.3 64.00 6.26 11226 9012
0.5 8.47 62.29 0 72.06 104 7.36 261.1 62.29 6.25 10903 8494
0.6 57.20 0 72.06 104 7.34 260.3 57.20 6.24 9937 6942
0.7 48.71 0 72.06 104 7.29 258.7 48.71 6.20 8328 4392
0.8 38.20 0 71.06 82.5 21.19 228.1 38.20 18.01 4965 1397
0.9 53.48 0 71.06 82.5 23.36 237.0 53.48 19.86 7822 5906
1.0 62.00 0 71.06 104 8.13 261.0 62.00 6.25 10848 10716
Note:
1. Prestressingsteel, f
ps
= f
pu
(1 k
c
dp
), k = 2(1.04
fpy
fpu
)
2. For anged section, c/d
e
0.42,
M
n
= A
ps
f
ps
(d
p

a
2
) +A
s
f
y
(d
s

a
2
)
A

y
(d

s

a
2
) +0.85f

c
(b b
w
)
1
h
f
(
a
2

h
f
2
)
a =
1
C
c =
Aps fpu+As fyA

s
f

y
0.85
1
f

c
(bbw)h
f
0.85
1
f

c
bw+kAps
fpu
dp
3. For anged section, c/d
e
> 0.42over-reinforced,
M
n
= (0.36
1
0.08
2
1
)f

c
b
w
d
2
e
+0.85
1
f

c
(b b
w
)h
f
(d
e
0.5h
f
)
d
e
=
Aps fps dp+As fyds
Aps fps +As fy
4. For rectangular section, i.e., when c < h
f
takeb = b
w
in theaboveformulas.
c 1999by CRCPressLLC
12. Shear Strength Design for Interior Girder, Strength Limit StateI
AASHTOrequiresthat for thestrength limit stateI
V
u
V
n
V
u
=

i
V
i
= 0.95
_
1.25(V
DC1
+V
DC2
) +1.5V
DW
+1.75V
LL+IM
_
+V
ps
where is shear resistance factor 0.9 and V
ps
is the secondary shear due to prestress.
Factored shear demands, V
u
, for the interior girder are calculated in Table 10.19. To
determinetheeffectivewebwidth, assumethat theVSL post-tensioningsystemof 5to12
tendon units[ 25] will beused with agrouted duct diameter of 2.88in. In thisexample,
b

= 12 2.88/2 = 10.56 in. (268 mm). Detailed calculationsof theshear resistance,

V
n
(using two-leg #5 stirrups, A

= 0.62 in.
2
[ 419 mm
2
] ) for span 1, are shown in
Table 10.20. The results for span 2 are similar to span 1 and the calculations are not
repeated for thisexample.
TABLE10.19 Factored Shear for an Interior Girder (Span 1)
V
DC1
V
DC2
V
DW
(kips) (kips) (kips) V
LL+IM
M
LL+IM
V
ps
M
u
Location Dead Dead Wearing (kips) (k-ft) (kips) V
u
(k-ft)
(x/L) load-1 load-2 surface Envelopes Associated P/S (kips) Associated
0.0 125.2 11.4 15.6 60.0 0 13.03 297.1 0
0.1 91.5 8.4 11.4 50.1 787 13.03 231.0 4017
0.2 57.7 5.3 7.2 42.0 1320 13.03 168.0 6883
0.3 24.0 2.2 3.0 34.3 1614 13.03 105.4 8472
0.4 9.7 0.9 1.2 27.7 1650 130.3 47.3 8903
0.5 43.4 4.0 5.4 35.1 1628 13.03 109.2 8457
0.6 77.1 7.1 9.6 42.0 1424 13.03 170.3 6929
0.7 111 10.1 13.8 49.9 852 13.03 233.1 4011
0.8 145 13.2 18.0 59.2 216 13.03 298.3 205.4
0.9 178 16.3 22.2 68.8 667 13.03 364 4790
1.0 216 19.4 26.4 78.5 1788 13.03 434.3 10191
Note: V
u
= 0.95[1.25
_
V
DC1
+V
DC2
_
+1.5V
DW
+1.75V
LL+IM
] +V
ps
10.A.2Three-Span, Continuous, Composite PlateGirder Bridge
Given: A three-span, continuous, compositeplategirder bridgehastwo equal spansof length 160ft
(48.8m) andonemidspan of 210ft (64m). Thesuperstructureis44ft (13.4m) wide. Theelevation,
plan, and typical cross-section areshown in Figure10.69.
Structural steel: A709Grade50for web and anges,
F
yw
= F
yt
= F
yc
= F
y
= 50 ksi (345 MPa)
A709 Grade 36 for stiffeners, etc.,
F
ys
= 36 ksi (248 MPa)
Concrete:
f

c
= 3250 psi (22.4 MPa),
E
c
= 3250 ksi (22,400 MPa),
modular ratio, n = 9
c 1999by CRCPressLLC
TABLE10.20 Shear Strength Design for Interior Girder, Strength Limit StateI (Span 1)
Location d

V
p

x
V
c
S V
n
|V
u
|
(x/L) (in.) (rad) (kips) /f

c
(1000) (degree) (kips) (in.) (kips) (kips)
0.0 54.00 0.084 124.1 0.090 0.028 23.5 6.50 234.4 12 460.7 297.1
0.1 54.00 0.063 93.9 0.071 0.093 27 5.60 201.8 12 400.4 231.0
0.2 52.90 0.042 63.1 0.055 0.733 33 2.37 83.7 24 194.0 168.0
0.3 58.87 0.021 31.8 0.034 1.167 38 2.10 82.5 24 167.6 105.4
0.4 60.84 0.000 0.0 0.020 1.078 36 2.23 90.6 24 150.2 47.3
0.5 59.14 0.018 27.8 0.037 1.026 36 2.23 88.0 24 171.0 109.2
0.6 54.06 0.036 56.2 0.058 0.539 30 2.48 89.5 24 196.4 170.3
0.7 54.00 0.054 83.5 0.077 0.106 27 5.63 202.9 12 392.0 233.1
0.8 54.00 0.072 110.4 0.097 0.287 23.5 6.50 234.3 12 448.4 298.3
0.9 54.00 0.090 136.8 0.117 0.137 23.5 3.49 125.8 9 420.5 364.0
1.0 57.42 0.000 0.0 0.199 2.677 36 1.0 38.3 3.5 478.9 434.3
Note:
1. b

= 10.56in. and y

isslopeof theprestressingcable.
2. A

= 0.62in.
2
(2#5)
V
n
= thelesser of
_
V
c
+V
s
+V
p
0.25f

c
b

+V
p
V
c
= 0.0316
_
f

c
b

, V
s
=
A fyd cos
s
=
VuVp
b d
,
x
Mu
d
+0.5Nu+0.5Vu cot Aps fpo
Es As +EpAps
0.002
F

=
Es As +EpAps
EcAc+Es As +EpAps
(when
x
isnegative, multiply by F

)
A
min
= 0.0316
_
f

c
b S
fy
For V
u
< 0.1f

c
b

, S
max
= smaller of
_
0.8d

24 in.
For V
u
0.1f

c
b

, S
max
= smaller of
_
0.4d

12 in.
Loads:
Dead load = self-weight + barrier rail + futurewearing3 in ACoverlay
Liveload = AASHTOHS20-44+ dynamicload allowance
Single-laneaveragedaily truck trafc(ADTT) = 3600 (oneway)
Deck: Concreteslab deck with thicknessof 10.875in. (276mm) hasbeen designed
Construction: Unshored; unbraced length for compression ange, L
b
= 20 ft (6.1m)
Specication: AASHTO-LRFD [ 1] (referred to asAASHTOin thisexample)
Requirements: Design the following portions of an interior girder for maximum positive exure
region at span 1:
1. Calculateloads
2. Calculateliveload distribution factors
3. Calculateunfactored momentsand shear demands
4. Determineload factorsfor strength limit stateI and fatiguelimit state
5. Calculatecompositesection propertiesfor positiveexureregion
6. Flexural strength design, strength limit stateI
7. Shear strength design, strength limit stateI
8. Fatiguedesign, fatigueand fracturelimit state
9. Intermediatetransversestiffener design
10. Shear connector design
11. Constructability check
c 1999by CRCPressLLC
FIGURE10.69: A three-span, continuousplategirder bridge.
Solution
1. CalculateLoads
1.1) Component DeadLoad, DC for anInterior Girder
Thecomponent dead load, DC, includesall structural dead loadswith theexception of
thefurther wearingsurfaceandspeciedutilityloads. For design purposes, thetwoparts
of DC aredened as:
DC1: Deck concrete (self-weight, 150 lb/ft
3
) and steel girder including bracing
systemanddetails(estimatedweight, 300lb/ft for eachgirder). Assumethat DC1 is
actingonthenoncompositesectionandisdistributedtoeachgirder bythetributary
area. Thetributary width for theinterior girder is16ft (4.9m).
c 1999by CRCPressLLC
DC1 = [(10.875/12)(16) +(1.5)(15.25 10.975)/12(1.5)] (0.15) +0.3
= 2.557 kips/ft (37.314 kN/m)
DC2: Barrier rail weight (784kips/ft). Assumethat DC2 isactingonthelong-term
compositesection and isequally distributed to each girder.
DC2 = 0.784/3 = 0.261 kips/ft (3.809 kN/m)
1.2) WearingSurfaceLoad, DW
A futurewearingsurfaceof 3in. (76mm) with aunit weight of 140lb/ft
3
isassumed to
becarried by thelong-termcompositesection and equally distributed to each girder.
DW = (deck widthbarrier width) (thicknessof wearingsurface)
(unit weight)/3
= [44 2(1.75)] (0.25)(0.14)/3 = 0.473 kips/ft (6.903 kN/m)
1.3) LiveLoad, LL, andDynamicLoadAllowance, IM
Thedesign liveload, LL, istheAASHTO HS20-44 vehicular liveload. To consider the
wheel loadimpact frommovingvehicles, thedynamicloadallowance, IM =33%for the
strengthlimit stateand15%for thefatiguelimit stateareused[ AASHTOTable3.6.2.1-1] .
2. CalculateLiveLoad Distribution Factors
2.1) RangeApplicabilityof AASHTOApproximateFormulas
AASHTO-LRFD [ 1] recommends that approximatemethods beused to distributelive
loadtoindividual girders. For concretedeck onsteel girders, liveloaddistributionfactors
aredependent onthegirder spacing, S, spanlength, L, concreteslabdepth, t
s
, longitudinal
stiffnessparameter, K
g
, andnumber of girders, N
b
. Therangeof applicabilityof AASHTO
approximateformulas are3.5 ft S 16 ft; 4.5 in. t
s
12 in.; 20 ft L 240
ft; and N
b
4. For this design example, S = 16 ft, L
1
= L
3
= 160 ft, L
2
= 210 ft,
t
s
= 10.875 in., and N
b
= 3 < 4. It is obvious that this bridgeis out of therangeof
applicability of AASHTOformulas. Theconventional level ruleisused to determinelive
load distribution factors.
2.2) Level Rule
Thelevel ruleassumesthat thedeck in itstransversedirection issimply supported by the
girdersand usesstaticsto determinetheliveload distribution to thegirders. AASHTO
also requiresthat when thelevel ruleisused, themultiplepresencefactor, m(1.2for one
loaded lane, 1.0for two loaded lanes, 0.85for threeloaded lanes, and 0.65for morethan
threeloaded lanes), should apply.
2.3) LiveLoadDistributionFactorsfor StrengthLimit State
Figure10.70 showslocationsof trafc lanesfor theinterior girder. For a12-ft (3.6-m)
trafclanewidth, thenumber of trafclanesfor thisbridgeisthree.
(a) Onelaneloaded (Figure10.70a)
R =
13
16
= 0.8125 lanes
LD = mR = 1.2(0.8125) = 0.975 lanes
c 1999by CRCPressLLC
FIGURE10.70: Liveload distributionlever rule.
(b) Two lanesloaded (Figure10.70b)
R =
13
16
+
9
16
= 1.375 lanes
LD = mR = 1.0(1.375) = 1.375 lanes (controls)
(c) Threelanesloaded (Figure10.70c)
R =
(13 +3)
16
+
7
16
= 1.4375 lanes
LD = mR = 0.85(1.4375) = 1.222 lanes
c 1999by CRCPressLLC
2.4) LiveLoadDistributionFactorsfor FatigueLimit State
AASHTOrequiresthat onetrafclaneload beused and multiplepresencefactorsnot be
applied to thefatiguelimit state. Theliveload distribution factor for thefatiguelimit
state, therefore, isobtained by onelaneloaded without amultiplepresencefactor of 1.2.
LD = 0.813
3. CalculateUnfactored Momentsand Shear Demands
For aninterior girder, unfactoredmoment andshear demandsareshowninFigures10.71
and 10.72 for the strength limit state and Figures 10.73 and 10.74 for the fatigue limit
state. Thedetailsarelisted in Tables10.21 to 10.23. Only theresultsfor span 1 and one
half of span 2areshown in thesetablesand guressincethebridgeissymmetrical about
thecenterlineof span 2.
4. DetermineLoad Factorsfor Strength Limit StateI and FatigueLimit State
4.1) General DesignEquation(AASHTOArticle1.3.2)

i
Q
i
R
n
(10.43)
where
i
areloadfactorsand resistancefactors; Q
i
representsforceeffectsor demands;
and R
n
is the nominal resistance. is a factor related to ductility, redundancy, and
operational importanceof that beingdesigned and isdened as:
=
D

I
0.95 (10.44)
where

D
=
_
1.05 for nonductilecomponentsand connections
0.95 for ductilecomponentsand connections
(10.45)

R
=
_
1.05 for nonredundant members
0.95 for redundant members
(10.46)

I
=
_

_
1.05 operationally important bridge
0.95 general bridge
only apply to strength
and extremeevent limit states
(10.47)
For thisbridge, thefollowingvaluesareassumed:
Ductility Redundancy Importance
Limit states
D

R

I

Strengthlimit state 0.95 0.95 1.05 0.95
Fatiguelimit state 1.0 1.0 1.0 1.0
4.2) LoadFactorsandLoadCombinations
Theload factorsand combinationsarespecied as(AASHTOTable3.4.1-1):
Strength limit stateI: 1.25(DC1 +DC2) +1.5(DW) +1.75(LL +IM)
Servicelimit state: 0.75(LL +IM)
5. CalculateCompositeSection Propertiesfor PositiveFlexureRegion
Try steel section (Figure10.75) as:
c 1999by CRCPressLLC
FIGURE10.71: Moment envelopesdueto unfactored loads.
FIGURE10.72: Shear envelopesdueto unfactored loads.
c 1999by CRCPressLLC
FIGURE10.73: Unfactored moment dueto fatigueloads.
Top ange: b
f c
= 18in. t
f c
= 1in.
Web: D = 96in. t
w
= 0.625in.
Bottomange: b
f t
= 18in. t
f t
= 1.75in.
5.1) EffectiveFlangeWidth(AASHTOArticle4.6.2.6)
For an interior girder, theeffectiveangewidth
b
eff
= thelesser of
_

_
L
eff
4
=
115(12)
4
= 345 in.
12t
s
+
b
f
2
= (12)(10.875) +18/2 = 140 in. (controls)
S = (16)(12) = 192 in.
whereL
eff
is theeffectivespan length and may betaken as theactual span length for
simplysupportedspansandthedistancebetweenthepointsof permanent loadinection
for continuousspans; b
f
isthetop angewidth of thesteel girder.
5.2) ElasticCompositeSectionProperties
For thetypical section (Figure10.75) in thepositiveexureregion of span 1, theelastic
section properties for the noncomposite, the short-term composite (n = 9), and the
long-termcomposite(3n =27), respectively, arecalculated in Tables10.24to 10.26.
5.3) PlasticMoment Capacity, M
p
Theplasticmoment capacity, M
p
, isdetermined usingequilibriumequations. Therein-
forcement in theconcreteslab isneglected in thisexample.
(a) Determinethelocation of theplasticneutral axis(PNA)
Assuming that thePNA is within thetop angeof thesteel girder (Figure10.76)
and that y
PNA
isthedistancefrom thetop of thecompression angeto thePNA,
weobtain:
c 1999by CRCPressLLC
TABLE10.21 Moment Envelopesfor Strength Limit StateI
M
DC1
M
DC2
M
DW
(kips-ft) (kips-ft) (kips-ft) M
LL+IM
M
u
Location Dead Dead Wearing (kips-ft) (kips-ft)
Span (x/L) load-1 load-2 surface Positive Negative Positive Negative
0.0 0 0 0 0 0 0 0
0.1 2047 209 379 1702 348 6049 2641
0.2 3439 351 636 2949 696 10310 4250
0.3 4177 426 773 3784 1042 12858 4835
0.4 4260 435 788 4202 1390 13684 4387
1 0.5 3688 376 682 4212 1738 12800 2908
0.6 2462 251 455 3829 2086 10236 402
0.7 582 59 108 3069 2434 6017 3131
0.8 1954 199 361 1951 2782 173 7696
0.9 5143 525 951 941 3736 6522 14297
1.0 8988 917 1663 637 5720 13074 23641
0.0 8988 917 1663 637 5720 13074 23641
0.1 3913 399 724 924 2998 4616 11136
0.2 33 3 6 2230 1695 3759 2767
2 0.3 2852 291 528 3499 1607 10302 1812
0.4 4544 464 841 4448 1607 14540 4473
0.5 5108 521 945 4766 1607 15954 5359
Note: Live load distribution factor, LD = 1.375. Dynamic load allowance, IM = 33%. M
u
=
0.95
_
1.25
_
M
DC1
+M
DC2
_
+1.5M
DW
+1.75M
LL+IM
_
TABLE10.22 Shear Envelopesfor Strength Limit StateI
V
DC1
V
DC2
V
DW
(kips-ft) (kips-ft) (kips-ft) V
LL+IM
V
u
Location Dead Dead Wearing (kips-ft) (kips-ft)
Span (x/L) load-1 load-2 surface Positive Negative Positive Negative
0.0 148.4 15.1 27.4 133.7 23.4 455.4 194.3
0.1 107.5 11.0 19.9 110.1 24.9 352.2 127.7
0.2 66.6 6.8 12.3 90.6 34.2 255.3 47.8
0.3 25.6 2.6 4.7 75.2 45.7 165.2 35.7
0.4 15.3 1.6 2.8 60.5 59.0 76.5 122.1
1 0.5 56.2 5.7 10.4 46.6 74.3 10.8 211.8
0.6 97.1 9.9 18.0 33.7 91.2 96.7 304.3
0.7 138.0 14.1 25.5 22.1 109.5 180.2 398.9
0.8 178.9 18.3 33.1 12.0 129.0 261.5 495.8
0.9 219.8 22.4 40.7 6.5 149.5 334.9 594.1
1.0 260.7 26.6 48.2 4.4 170.5 402.5 693.3
0.0 268.5 27.4 49.7 181.6 15.0 724.2 397.3
0.1 214.7 21.9 39.7 154.4 15.8 594.2 311.2
0.2 161.1 16.4 29.8 128.3 22.1 466.5 216.4
2 0.3 107.4 11.0 19.9 104.0 32.5 341.8 115.0
0.4 53.7 5.5 9.9 81.8 45.9 220.4 8.1
0.5 0 0 0 62.4 62.4 103.8 103.8
Note: Live load distribution factor, LD = 1.375. Dynamic load allowance, IM = 33%. V
u
=
0.95
_
1.25
_
V
DC1
+V
DC2
_
+1.5V
DW
+1.75V
LL+IM
_
P
s
+P
c1
= P
c2
+P
w
+P
t
(10.48)
where
P
s
= 0.85f

c
b
eff
t
s
= 0.85(3.25)(140)(10.875) = 4206 kips(18,708kN)
P
c1
= y
PNA
b
f c
F
yc
P
c2
= A
f c
F
yc
P
c1
= (t
f c
y
PNA
)b
f c
F
yc
P
c
= P
c1
+P
c2
= A
f c
F
yc
= (18)(1)(50) = 900 kips(4,003kN)
P
w
= A
w
F
yw
= (96)(0.625)(50) =3,000kips(13,344kN)
P
t
= A
f t
F
yt
= (18)(1.75)(50) =1,575kips(7,006kN)
Substitutingtheaboveexpressionsinto Equation (10.48) and solvingfor y
PNA
, we
obtain
c 1999by CRCPressLLC
FIGURE10.74: Unfactored shear dueto fatigueloads.
FIGURE10.75: Typical cross-section in positiveexureregion.
y
PNA
=
t
f c
2
_
P
w
+P
t
P
s
P
c
+1
_
(10.49)
y
PNA
=
1
2
_
3000 +1575 4206
900
+1
_
= 0.705 in. < t
cf
= 1.0 in. O.K.
(b) CalculateM
p
c 1999by CRCPressLLC
TABLE10.23 Moment and Shear Envelopesfor FatigueLimit State
M
LL+IM
V
LL+IM
(M
LL+IM
)
u
(V
LL+IM
)
u
Location (kips-ft) (kips) (kips-ft) (kips)
Span (x/L) Positive Negative Positive Negative Positive Negative Positive Negative
0.0 0 0 68.2 11.9 0 0 51.1 8.9
0.1 868 177 56.2 12.7 651 133 42.1 9.5
0.2 1504 355 46.2 17.5 1128 266 34.7 13.1
0.3 1930 532 38.4 23.3 1447 399 28.8 17.5
0.4 2143 709 30.9 30.1 1607 532 23.1 22.6
1 0.5 2148 886 23.8 37.9 1611 665 17.8 28.4
0.6 1953 1064 17.2 46.5 1465 798 12.9 34.9
0.7 1565 1241 11.3 55.8 1174 931 8.5 41.9
0.8 995 1419 6.1 65.8 746 1064 4.6 49.3
0.9 480 1905 3.3 76.2 360 1429 2.5 57.2
1.0 325 2917 2.2 87.0 243 2188 1.7 65.2
0.0 325 2917 92.6 7.6 243 2188 69.5 5.7
0.1 471 1529 78.7 8.1 353 1146 59.1 6.0
0.2 1137 865 65.4 11.3 853 648 49.1 8.5
2 0.3 1785 820 53.0 16.5 1338 615 39.8 12.4
0.4 2268 820 41.7 23.4 1701 615 31.3 17.6
0.5 2430 820 31.8 31.8 1823 615 23.9 23.9
Note: Live load distribution factor, LD = 0.813. Dynamic load allowance, IM = 15%.
_
M
LL+IM
_
u
=
0.75
_
M
LL+IM
_
u
and
_
V
LL+IM
_
u
= 0.75
_
V
LL+IM
_
u
TABLE10.24 NoncompositeSection Propertiesfor PositiveFlexureRegion
A y
i
A
i
y
i
y
i
y
sb
A
i
_
y
i
y
sb
_
2
I
o
Component (in.
2
) (in.) (in.
3
) (in.) (in.
4
) (in.
4
)
Top ange, 18x 1 18 98.25 1,768.5 54.587 53,636 1.5
Web, 96x 0.625 60 49.75 2,985.0 6.087 2,223 46,080
Bottomange, 18x 1.75 31.5 0.875 27.6 42.788 57,670 8.04

109.5 4,781.1 113,529 46,090

y
sb
=

A
i
y
i

A
i
=
4,781.1
109.5
= 43.663 in.
y
st
= (1.75 +96 +1) 43.663 = 55.087 in.
I
girder

I
o
+

A
i
_
y
i
y
sb
_
2
= 46,090 +113,529 = 159,619 in.
4
S
sb
=
I
girder
y
sb
=
159,619
43.663
= 3,656 in.
3
S
st
=
I
girder
y
st
=
159,619
55.087
= 2,898 in.
3
Summingall forcesabout thePNA, weobtain:
M
p
=

M
PNA
= P
s
d
s
+P
c1
_
y
PNA
2
_
+P
c2
_
t
f c
y
PNA
2
_
+P
w
d
w
+P
t
d
t
(10.50)
where
d
s
=
10.875
2
+4.375 1 +0.705 = 9.518 in. (242 mm)
d
w
=
96
2
+1 0.705 = 48.295 in. (1,227 mm)
d
t
=
1.75
2
+96 +1 0.705 = 97.17 in. (2,468 mm)
M
p
= (4,206)(9.518) +(18)(50)
(0.705)
2
2
+(18)(50)
(10.705)
2
2
+(3,000)(48.295) +(1,575)(97.17)
c 1999by CRCPressLLC
TABLE10.25 Short-TermCompositeSection Properties(n =9)
A y
i
A
i
y
i
y
i
y
sbn
A
i
_
y
i
y
sbn
_
2
I
o
Component (in.
2
) (in.) (in.
3
) (in.) (in.
4
) (in.
4
)
Steel section 109.5 43.663 4,781.1 38.791 164,768 159,619
Concreteslab
140/9x 10.875 169.17 107.563 18,196 25.109 106,653 1,667

278.67 22,977 271,421 161,286

y
sbn

A
i
y
i

A
i
=
22,977
278.67
= 82.454 in.
y
st n
= (1.75 +96 +1) 82.454 = 16.296 in.
I
comn
=

I
o
+

A
i
_
y
i
y
sbn
_
2
= 161,286 +271,421 = 432,707 in.
4
S
sbn
=
I
conn
y
sbn
=
432,707
82.454
= 5,248 in.
3
S
st n
=
I
comn
y
st n
=
432,707
16.296
= 26,553 in.
3
TABLE10.26 Long-TermCompositeSection Properties(3n =27)
A y
i
A
i
y
i
y
i
y
sb3n
A
i
_
y
i
y
sb3n
_
2
I
o
Component (in.
2
) (in.) (in.
3
) (in.) (in.
4
) (in.
4
)
Steel section 109.5 43.663 4,781.1 21.72 51,661 159,619
Concreteslab
140/9x 10.875 56.39 107.563 6,065.4 42.18 100,320 556

165.89 10,846.4 151,981 160,174

y
sb3n
=

A
i
y
i

A
i
=
10,846.4
165.89
= 65.383 in.
y
st 3n
= (1.75 +96 +1) 65.383 = 33.367 in.
I
com3n
=

I
o
+

A
i
_
y
i
y
sb3n
_
2
= 160,174 +151,981 = 312,155 in.
4
S
sb3n
=
I
con3n
y
sb3n
=
312,155
65.383
= 4,774 in.
3
S
st 3n
=
I
com3n
y
st 3n
=
312,155
33.367
= 9,355 in.
3
= 338,223 kips-in. = 28,185 kips-ft (38,212 kN-m)
5.4) YieldMoment, M
y
(AASHTOArticle6.10.5.1.2)
The yield moment, M
y
, corresponds to the rst yielding of either steel ange. It is
obtained by thefollowingformula:
M
y
= M
D1
+M
D2
+M
AD
(10.51)
whereM
D1
, M
D2
, and M
AD
aremomentsdueto thefactored loadsapplied to thesteel,
thelong-termand theshort-termcompositesection, respectively. M
AD
can beobtained
by solvingequation:
c 1999by CRCPressLLC
FIGURE10.76: Plasticmoment capacity state.
F
y
=
M
D1
S
s
+
M
D2
S
3n
+
M
AD
S
n
(10.52)
M
AD
= S
n
_
F
y

M
D1
S
s

M
D2
S
3n
_
(10.53)
(10.54)
whereS
s
, S
n
, andS
3n
(seeTables10.24to10.26) aresectionmoduli for thenoncomposite
steel, theshort-termandthelong-termcompositesection, respectively. FromTable10.21,
maximumfactoredpositivemomentsM
D1
andM
D2
inspan1areobtainedat thelocation
of 0.4L
1
.
M
D1
= (0.95)(1.25)(M
DC1
) = (0.95)(1.25)(4260) = 5,059 kips-ft
M
D2
= (0.95) (1.25M
DC2
+1.5M
DW
)
= (.095) [1.25(435) +1.5(788)] = 1,640 kips-ft
For thetop ange:
M
AD
= (26,553)
_
50
5,059(12)
2,898

1,640(12)
9,355
_
= 715,552 kips-in. = 59,629 kips-ft (80,842 kN-m)
For thebottomange:
c 1999by CRCPressLLC
M
AD
= (5,248)
_
50
5,059(12)
3,656

1,640(12)
4,774
_
= 153,623 kips-in. = 12,802 kips-ft (17,356 kN-m) (controls)
.
.. M
y
= 5,059 +1,640 +12,802 = 19,501 kips-ft (26,438 kN-m)
6. Flexural Strength Design, Strength Limit StateI
6.1) Compactnessof Steel Girder Section
Thesteel sectionisrst checkedtomeet therequirementsof acompact section(AASHTO
Article6.10.5.2.2).
(a) Ductility requirement:
D
p

d +t
s
+t
h
7.5
whereD
p
isthedepthfromthetopof theconcretedeck tothePNA, d isthedepthof
thesteel girder, andt
h
isthethicknessof theconcretehaunchabovethetopangeof
thesteel girder. Thepurposeof thisrequirement isto prevent permanent crashing
of the concrete slab when the composite section approaches its plastic moment
capacity. For thisexample, referringto Figure10.75and 10.76, weobtain:
D
p
= 10.875 +4.375 1 +0.705 = 14.955 in. (381 mm)
D
p
= 14.955 in. <
d +t
s
+t
h
7.5
=
98.75 +10.875 +3.375
7.5
= 15.067 in. O.K.
(b) Web slendernessrequirement,
2D
cp
t
w
3.76
_
E
F
yc
where D
cp
is the depth of the web in compression at the plastic moment state.
SincethePNA iswithin thetop ange, D
cp
isequal to zero. Theweb slenderness
requirement issatised.
(c) Compression angeslendernessand compression angebracingrequirement
It isusuallyassumedthat thetopangeisadequatelybracedbythehardenedconcrete
deck; thereare, therefore, no requirementsfor thecompression angeslenderness
and bracingfor compact compositesectionsat thestrength limit state.
.
.. thesection isacompact compositesection.
6.2) Moment of InertiaRatioLimit (AASHTOArticle6.10.1.1)
Theexural membersshall meet thefollowingrequirement:
0.1
I
yc
I
y
0.9
whereI
yc
and I
y
arethemomentsof inertiaof thecompression angeand steel girder
about thevertical axisin theplaneof web, respectively. Thislimit ensuresthat thelateral
torsional buckingformulasarevalid.
c 1999by CRCPressLLC
I
yc
=
(1)(18)
3
12
= 486 in.
4
I
y
= 486 +
(96)(0.625)
3
12
+
(1.75)(18)
3
12
+1338 in.
4
0.1 <
I
yc
I
y
=
486
1338
= 0.36 < 0.9 O.K.
6.3) Nominal FlexureResistance, M
n
(AASHTOArticle10.5.2.2a)
It isassumedthat theadjacent interior-pier sectionisnoncompact. For continuousspans
withthenoncompact interior support section, thenominal exureresistanceof acompact
compositesection istaken as:
M
n
= 1.3R
h
M
y
M
p
(10.55)
whereR
h
isaangestressreduction factor taken as1.0for thishomogeneousgirder.
M
n
= 1.3(1.0)(19,501) = 25,351 kips-ft < M
p
= 28,185 kips-ft
6.4) StrengthLimit StateI
AASHTO-LRFD [ 1] requiresthat for strength limit stateI
M
u

f
M
n
(10.56)
where
f
istheexural resistancefactor =1.0. For thecompositesection in thepositive
exureregion in span 1, themaximummoment occursat 0.4L
1
(seeTable10.21).
M
u
= 13,684 kips-ft <
f
M
n
= (1.0)(25,351) = 25,351 kips-ft O.K.
7. Shear Strength Design, Strength Limit StateI
7.1) Nominal Shear Resistance, V
n
(a) V
n
for an unstiffened web (AASHTOArticle6.10.7.2)
V
n
=
_

_
V
p
= 0.58F
yw
Dt
w
For
D
t
w
2.46
_
E
F
yw
1.48t
2
w
_
EF
yw
For 2.46
_
E
F
yw
<
D
t
w
3.07
_
E
F
yw
4.55t
3
w
E
D
For
D
t
w
> 3.07
_
E
F
yw
(10.57)
whereD isdepth of web and t
w
isthicknessof web.
..
.
D
t
w
=
96
0.625
= 153.6 > 3.07
_
E
F
yw
3.07
_
29,000
50
= 73.9
.
.. V
n
=
4.55t
3
w
E
D
=
4.55(0.625)
3
(29,000)
96
= 335.6 kips (1,493 kN)
(b) V
n
for an end-stiffened web panel (AASHTOArticle6.10.7.3.3c)
c 1999by CRCPressLLC
V
n
= CV
p
(10.58)
C =
_

_
1.0 For
D
t
w
< 1.10
_
Ek
F
yw
1.10
(D/t
w
)
_
Ek
F
yw
For 1.10
_
Ek
F
yw

D
t
w
1.38
_
Ek
F
yw
1.52
(D/t
w
)
2
_
Ek
F
yw
For
D
t
w
> 1.38
_
Ek
F
yw
(10.59)
k = 5 +
5
(d
o
/D)
2
(10.60)
in which d
o
isthespacingof transversestiffeners(Figure10.77).
FIGURE10.77: Typical steel girder dimensions.
For d
o
=240in. and k = 5 +
5
(240/96)
2
= 5.80
..
.
D
t
w
= 153.6 > 1.38
_
Ek
F
yw
= 1.38
_
29,000(5.8)
50
= 80
.
.. C =
152
(153.6)
2
=
_
29,000(5.80)
50
= 0.374
V
p
= 0.58F
yw
Dt
w
= 0.58(50)(96)(0.625) = 1,740 kips (7,740 kN)
V
n
= CV
p
= 0.374(1740) = 650.8 kips (2,895 kN)
c 1999by CRCPressLLC
(c) V
n
for interior-stiffened web panel (AASHTOArticle6.10.7.3a)
V
n
=
_

_
V
p
_
C +
0.87(1C)

1+(d
o
/D)
2
_
For M
u
0.5
f
M
p
RV
p
_
C +
0.87(1C)

1+(d
o
/D)
2
_
CV
p
For M
u
> 0.5
f
M
p
(10.61)
where
R =
_
0.6 +0.4
_

f
M
n
M
u

f
M
n
0.75
f
M
y
__
1.0 (10.62)
7.2) StrengthLimit StateI
AASHTO-LRFD [ 1] requiresthat for strength limit stateI
V
u

V
n
(10.63)
where

istheshear resistancefactor =1.0.

(a) Left end of span 1:
..
.
V
u
= 445.4 kips >

V
n
(for unstiffened web) = 335.6 kips
.
.. Stiffenersareneeded to increaseshear capacity.
In order to facilitatehandlingof web panel sections, thespacingof transversestiff-
enersshall meet (AASHTOArticle6.10.7.3.2) thefollowingrequirement:
d
o
D
_
260
(D/t
w
)
_
2
(10.64)
Try d
o
=240in. for end-stiffened web panel
d
o
= 240 in. < D
_
260
(D/t
w
)
_
2
= 96
_
260
96/0.625
_
2
= 275 in. O.K.
and then

V
n
= (1.0)650.8 = 650.8 kips > V
u
= 445.4 kips O.K.
(b) Location of therst intermediatestiffeners, 20ft (6.1m) fromtheleft endin span 1:
Factored shear for this location can be obtained using linear interpolation from
Table 10.22. Since V
u
= 328.0 kips (1459 kN) is less than the shear capacity of
the unstiffened web,

V
n
= 335.5 kips (1492 kN), the intermediate transverse
stiffenersmaybeomitted after therst intermediatestiffeners. Similar calculations
can beused to determinetheremainingstiffenersalongthegirder.
8. FatigueDesign, Fatigueand FractureLimit State
Thebasemetal at theconnectionplateweldstoanges, andwebslocatedat 96ft (29.26m)
(0.6L1) fromtheleft end of span 1will bechecked for thefatigueload combination.
c 1999by CRCPressLLC
8.1) Load-InducedFatigue(AASHTOArticle6.6.1.2)
Thedesign requirementsfor load-induced fatigueapply only to (1) detailssubjected to a
net applied tensilestressand (2) regionswheretheunfactored permanent loadsproduce
compression, and only if thecompressivestressislessthan twicethemaximum tensile
stressresultingfrom thefatigueload combination. In thefatiguelimit state, all stresses
arecalculated usingtheelasticsection properties(Tables10.24to 10.26).
(a) Top-angeweld
Thecompressivestressat thetop-angeweld duetounfactored permanent loadsis
obtained:
f
DC
=
M
DC1
(y
st
t
f c
)
I
girder
+
(M
DC2
+M
DW
)(y
st
t
f c
)
I
com3n
=
2462(12)(55.087 1.0)
159,619
+
(251 +455)(12)(33.367 1.0)
312,155
= 10.89 ksi (75.09 MPa)
Assumethat thenegativefatiguemomentsarecarried by thesteel section only in
thepositiveexureregion. Themaximum tensilestressat thetop-angeweld at
thislocation dueto factored fatiguemoment is
f
LL+IM
=

(M
LL+IM
)
u

_
y
st
t
f c
_
I
girder
=
798(12)(54.087)
159,619
= 3.25 ksi (22.41 MPa)
..
.
f
DC
= 10.89 ksi > 2f
LL+IM
= 6.49 ksi
.
.. no need to check fatiguefor thetop-angeweld
(b) Bottom-angeweld
Factored fatiguestressrange, (f )
u
For thepositiveexureregion, weassumethat positivefatiguemomentsare
applied totheshort-termcompositesection and negativefatiguemomentsare
applied to thenoncompositesteel section only.
(f )
u
=
(M
LL+IM
)
u
_
y
sbn
t
f t
_
I
comn
+

(M
LL+IM
)
u

_
y
sb
t
f t
_
I
girder
=
1465(12)(82.454 1.75)
432,707
+
798(12)(43.663 1.75)
159,619
= 5.79 ksi (39.92 MPa)
Nominal fatigueresistancerange, (F)
n
For let-welded connectionswith weld linesnormal to thedirection of stress,
thebasemetal at transversestiffenersto angeweldsisfatiguedetail category
C

(AASHTOTable6.6.1.2.3.-1).
c 1999by CRCPressLLC
(F)
n
=
_
A
N
_
1/3

1
2
(F)
T H
(10.65)
whereA isaconstant dependent on detail category = 44(10)
8
for category C

and
N = (365)(75)n(ADT T )
ST
(10.66)
ADT T
ST
= p(ADT T ) (10.67)
wherep isafraction of atruck in asinglelane(AASHTOTable3.6.1.4.2-1) =
0.8 for three-lanetrafc, and n isthenumber of stress-rangecyclesper truck
passage(AASHTOTable6.6.1.2.5-2) =1.0for thepositiveexureregion.
N = (365)(75)(1.0)(0.8)(3600) = 7.844(10)
7
For category C

detail, (F)
T H
=12ksi (AASHTOTable6.6.1.2.5-3).
..
.
_
A
N
_
1/3
=
_
44(10)
8
7.844(10)
7
_
1/3
= 3.83 ksi <
1
2
(F)
T H
= 6 ksi
.
.. (F)
n
=
1
2
(F)
T H
= 6 ksi (41.37 MPa)
Fatiguelimit state
AASHTOrequiresthat each detail shall satisfy:
(f )
u
(F)
n
(10.68)
For top-angeweld
(f )
u
= 5.79 ksi < (F)
n
= 6 ksi O.K.
8.2) FatigueRequirementsfor Web(AASHTOArticle6.10.4)
Thepurposeof theserequirementsisto control out-of-planeexing of theweb dueto
exureand shear under repeated liveloadings. Therepeated liveload istaken astwice
thefactored fatigueload.
(a) Flexurerequirement
f
cf

_

_
R
h
F
yc
For
2Dc
tw
5.76
_
E
Fyc
R
h
F
yc
_
3.58 0448
2Dc
tw
_
Fyc
E
_
For 5.76
_
E
Fyc

2Dc
tw
6.43
_
E
Fyc
28.9R
h
E
_
tw
2Dc
_
2
For
2Dc
tw
> 6.43
_
E
Fyc
(10.69)
wheref
cf
isthemaximum elastic exural stressin thecompression angedueto
theunfactoredpermanent loadsandrepeatedliveloadings; F
yc
istheyieldstrength
of thecompression ange; and D
c
isthedepth of theweb in compression.
Depth of web in compression, Dc
Consideringthealgebraicsumof stressesactingon different sectionsbased on
elasticsection properties, D
c
can beobtained by thefollowingformula:
c 1999by CRCPressLLC
D
c
=
f
DC1
+f
DC2
+f
DW
+f
LL+IM
f
DC1
y
st
+
f
DC2
+f
DW
y
st 3n
+
f
LL+IM
y
st n
t
f c
=
M
DC1
S
st
+
M
DC2
+M
DW
S
st 3n
+
2(M
LL+IM
)u
S
st n
M
DC1
I
girder
+
M
DC2
+M
DW
I
com3n
+
2(M
LL+IM
)u
I
comn
t
f c
(10.70)
Substituting moments (Tables 10.21 and 10.23) and section properties (Ta-
bles10.24and 10.26) into Equation 10.70, weobtain:
D
c
=
4260(12)
2,898
+
(435+788)(12)
9,355
+
2(1607)(12)
26,553
4260(12)
159,629
+
(435+788)(12)
312,155
+
2(1607)(12)
432,707
1
=
17.640 +1.569 +1.452
0.320 +0.047 +0.089
1 = 44.29 in. (1,125 mm)
2D
c
t
w
=
2(44.29)
0.625
= 141.7 < 5.76
_
E
F
yc
= 183.7
Maximumcompressivestressin ange, f
cf
(at location 0.4L
1
)
f
cf
= f
DC1
+f
DC2
+f
DW
+f
LL+IM
=
M
DC1
S
st
+
M
DC2
+M
DW
S
st 3n
+
2(M
LL+IM
)
u
S
st n
= 17.64 +1.57 +1.45 = 20.66 < R
h
F
yc
= 50 ksi
(b) Shear (AASHTOArticle10.6.10.4.4)
Theleft end of span 1ischecked asfollows:
Fatigueload
V
u
= V
DC1
+V
DC2
+V
DW
+2(V
LL+IM
)
u
= 148.4 +15.1 +27.4 +2(51.1) = 293.1 kips (1304 kN)
Fatigueshear stress

cf
=
V
u
Dt
w
=
293.1
96(0.625)
= 4.89 ksi (33.72 MPa)
Fatigueshear resistance
C = 0.374 (seeStep 7)

n
= 0.58CF
yw
= 0.58(0.374)(50)
= 10.85 ksi >
cf
= 4.89 ksi O.K.
c 1999by CRCPressLLC
8.3) Distortion-InducedFatigue(AASHTOArticle6.6.1.3)
All transverse connection plates will be welded to both the tension and compression
anges to provide rigid load paths so distortion-induced fatigue (the development of
signicant secondary stresses) can beprevented.
8.4) FractureLimit State(AASHTOArticle6.6.2)
Materialsfor main load-carryingcomponentssubjected to tensilestresseswill meet the
Charpy V-notch fracturetoughnessrequirement (AASHTOTable6.6.2-2) for tempera-
turezone2(AASHTOTable6.6.2-1).
9. IntermediateTransverseStiffener Design
Theintermediatetransversestiffener consists of two plates welded to both sides of the
web. Thedesign of therst intermediatetransversestiffener isdiscussedin thefollowing.
9.1) ProjectingWidth, b
t
, Requirements(AASHTOArticle6.10.8.1.2)
Toprevent local bucklingof thetransversestiffeners, thewidthof eachprojectingstiffener
shall satisfy theserequirements:
_
2.0 +
d
30
0.25b
f
_
b
t

_
0.48t
p
_
E
F
ys
16t
p
_
(10.71)
whereb
f
isthefull widthof thesteel angeandF
ys
isthespeciedminimumyieldstrength
of thestiffener. To allowadequatespacefor cross-frameconnections, try stiffener width
b
t
=6in. (152mm):
b
t
= 6 in. >
_
2.0 +
d
30
= 2.0 +
98.75
30
= 5.3 in.
0.25b
f
= 0.25(18) = 4.5 in. O.K.
Try t
p
=0.5in. (13mm) and obtain:
b
t
= 6 in. <
_
0.48t
p
_
E
F
ys
= 0.48(0.5)
_
29,000
36
= 6.8 in.
16t
p
= 16(0.5) = 8 in. O.K.
Usetwo 6in. x 0.5in. (152mmx 13mm) transversestiffener plates.
9.2) Moment of InertiaRequirement (AASHTOArticle6.10.8.1.3)
Thepurposeof thisrequirement isto ensuresufcient rigidity of transversestiffenersto
adequately develop atension eld in theweb.
I
t
d
o
t
2
w
J (10.72)
J = 2.5
_
D
p
d
o
_
2
2.0 0.5 (10.73)
whereI
t
isthemoment of inertiafor thetransversestiffener takenabout theedgeincontact
withthewebfor singlestiffenersandabout themid-thicknessof thewebfor stiffener pairs
(Figure10.78); D
p
istheweb depth for webswithout longitudinal stiffeners.
c 1999by CRCPressLLC
FIGURE10.78: Cross-section of web and transversestiffener.
..
.
J = 2.5
_
96
240
_
2
2.0 = 1.6 < 0.5
..
.
Use J = 0.5
I
t
= 2
_
6
3
(0.5)
12
+(6)(0.5)(3.313)
2
_
= 83.86 in.
4
> d
o
t
2
w
J = (240)(0.625)
2
(0.5) = 46.88 in.
4
O.K.
9.3) AreaRequirement (AASHTOArticle6.10.8.1.4)
Thisrequirement ensuresthat transversestiffenershavesufcient areatoresist thevertical
component of thetension eld, and is only applied to transversestiffeners required to
carry theforcesimposed by tension-eld action.
A
s
A
s min
=
_
0.15BDt
w
(1 C)
V
u

V
n
18t
2
w
__
F
yw
F
ys
_
(10.74)
whereB =1.0for stiffener pairs. Fromthepreviouscalculation:
C = 0.374, F
yw
= 50 ksi, F
ys
= 36 ksi
V
u
= 328.0 kips,
f
V
n
= 335.5 kips, t
w
= 0.625 in.
A
s
= 2(6)(0.5) = 6 in.
2
> A
s min
=
_
0.15(1.0)(96)(0.625)(1 0374)
328.0
335.5
18(0.625)
2
__
50
36
_
= 0.635 in.
2
c 1999by CRCPressLLC
Thenegativevalueof A
s min
indicatesthat thewebhassufcient areatoresist thevertical
component of thetension eld.
10. Shear Connector Design
In acompositegirder, studor channel shear connectorsmust beprovidedat theinterface
between theconcretedeck slab and thesteel section to resist theinterfaceshear. For a
compositebridgegirder, theshear connectorsshould benormally provided throughout
thelength of thebridge(AASHTOArticle6.10.7.4.1). Stud shear connectorsarechosen
in thisexampleand will bedesigned for thefatiguelimit stateand then checked against
the strength limit state. The detailed calculations of the shear stud connectors for the
positiveexureregion of span 1 aregiven in thefollowing. A similar procedurecan be
used to design theshear studsfor other portionsof thebridge.
10.1) StudSize(AASHTOArticle6.10.7.4.1a)
To meet thelimitsfor cover and penetration for shear connectorsspecied in AASHTO
Article6.10.7.4.1d, try:
Stud height, H
stud
= 7 in. > t
h
+2 = 3.375 +2 = 5.375 in. O.K.
Stud diameter, d
stud
= 0.875 in. < H
stud
/4 = 1.75 in. O.K.
10.2) Pitchof Shear Stud, p, for FatigueLimit State
(a) Basicrequirements(AASHTOArticle6.10.7.4.1b)
6d
stud
p =
n
stud
Z
r
I
comn
V
sr
Q
24 in. (10.75)
where n
stud
is the number of shear connectors in a cross-section; Q is the rst
moment of transformedsection (concretedeck) about theneutral axisof theshort-
term compositesection; V
sr
istheshear forcerangein thefatiguelimit state; and
Z
r
istheshear fatigueresistanceof an individual shear connector.
(b) Fatigueresistance, Z
r
(AASHTOArticle6.10.7.4.2)
Z
r
= d
2
stud
5.5d
2
stud
(10.76)
= 34.5 4.28 log N (10.77)
where N is the number of cycles specied in AASHTO Article 6.6.1.2.5, N =
7.844(10)
7
cycle(seeStep 8).
= 34.5 4.28 log(7.844 10
7
) = 0.711 < 5.5
Z
r
= 5.5d
2
stud
= 5.5(0.875)
2
= 4.211 ksi
(c) First moment, Q, and moment of initial, I
comn
(seeTable10.25)
c 1999by CRCPressLLC
Q =
_
b
eff
t
s
9
__
y
st n
t
h
+
t
s
2
_
=
_
140(10.875)
9
__
16.296 +3.375 +
10.875
2
_
= 4247.52 in.
3
I
comn
= 432,707 in.
4
(d) Required pitch for thefatiguelimit state
Assumethat shear studsarespaced at 6 in. transversely acrossthetop angeof a
steel section (Figure10.75) and, usingn
stud
=3for thisexample, obtain
P
required
=
3(4.211)(432,707)
V
sr
(4,247.52)
=
1,286.96
V
sr
The detailed calculations for the positive exure region of span 1 are shown in
Table10.27.
TABLE10.27 Shear Connector Design for
thePositiveFlexureRegion in Span 1
Location V
sr
P
required
P
nal
(x/L) (kips) (in.) (in.) n
total-stud
0.0 60.1 21.4 12 3
0.1 51.6 24.9 12 51
0.2 47.8 26.9 18 99
0.3 46.2 27.9 18 132
0.4 45.7 28.2 18 165
0.4 45.7 28.2 12 162
0.5 46.2 27.9 12 114
0.6 47.8 26.9 12 66
0.7 50.3 25.6 9 3
Note:
V
sr
=

+
_
V
LL+IM
_
u

_
V
LL+IM
_
u

P
required
=
n
stud
Zr I
comn
Vsr Q
=
1286.96
Vsr
n
total-stud
is the summation of number of shear studs be-
tween thelocationsof thezero moment and that location.
10.3) StrengthLimit StateCheck
(a) Basicrequirement (AASHTOArticle6.10.7.4.4a)
Theresulting number of shear connectors provided between thesection of max-
imum positivemoment and each adjacent point of zero moment shall satisfy the
followingrequirement:
n
total-stud

V
h

sc
Q
n
(10.78)
where
sc
is the resistance factor for shear connectors, 0.85; V
h
is the nominal
horizontal shear force; and Q
n
is the nominal shear resistance of one stud shear
connector.
(b) Nominal horizontal shear force(AASHTOArticle6.10.7.4.4b)
c 1999by CRCPressLLC
V
h
= thelesser of
_
0.85f

c
b
eff
t
s
F
yw
Dt
w
+f
yt
b
f t
t
f t
+F
yc
b
f c
t
f c
(10.79)
V
hconcrete
= 0.5f

c
b
eff
t
s
= 0.85(3.25)(140)(10.875) = 4,206 kips
V
hsteel
= F
yw
Dt
w
+F
yt
b
f t
t
f t
+F
yc
b
f c
t
f c
= 50 [(18)(1.0) +(96)(0.625) +(18)(1.75)] = 5,475 kips
.
.. V
h
= 4,206 kips (18,708 kN)
(c) Nominal shear resistance(AASHTOArticle6.10.7.4.4c)
Q
n
= 0.5A
sc
_
f

c
E
c
A
sc
F
u
(10.80)
whereA
sc
isacross-sectional areaof astud shear connector and F
u
isthespecied
minimumtensilestrength of astud shear connector =60ksi (420MPa).
..
.
0.5
_
f

c
E
c
= 0.5
_
3.25(3,250) = 51.4 kips < F
u
= 60 kips
.
.. Q
n
= 0.5A
sc
_
f

c
E
c
= 51.4
_
(0.875)
2
4
_
= 30.9 kips
(d) Check resultingnumber of shear stud connectors(seeTable10.27)
n
total-stud
=
_
165 fromleft end 0.4L
1
162 from0.4L
1
to 0.7L
1
_
>
V
h

sc
Q
n
=
4206
0.85(30.9)
= 160 O.K.
11. Constructability Check
For unshored construction, AASHTO requires that all I-section bending members be
investigated for strength and stability duringconstruction stagesusingappropriateload
combinationsgiven in AASHTO Table3.4.1-1. Thefollowing checksaremadefor the
steel girder section only under factored dead load, DC1. It isassumed that thenal total
dead load, DC1, producesthecontrollingmaximummoments.
11.1) WebSlendernessRequirement (AASHTO6.10.10.2.2)
2D
c
t
w
6.77
_
E
f
c
(10.81)
wheref
c
isthestressin compression angedueto thefactored dead load, DC1, and D
c
isthedepth of theweb in compression in theelasticrange.
c 1999by CRCPressLLC
D
c
= y
st
t
f c
= 55.087 1 = 54.087 in. (1,374 mm)
f
c
=
(0.95)(1.25)M
DC1
S
st
=
0.95(1.25)(4260)(12)
2,898
= 20.95 ksi (145 MPa)
2D
c
t
w
=
2(54.087)
0.625
= 173.1 6.77
_
E
f
c
= 6.77
_
29,000
20.95
= 251.9 O.K.
.
.. no longitudinal stiffener isrequired
11.2) CompressionFlangeSlendernessRequirement (AASHTOArticle6.10.10.2.3)
Thisrequirement preventsthelocal buckling of thetop angebeforetheconcretedeck
hardens.
b
f
2t
f
1.38

_
E
f
c
_
2D
c
t
w
(10.82)
b
f
2t
f
=
18
2(1.0)
= 9 1.38

_
E
f
c
_
2D
c
t
w
= 1.38
_
29,000
20.95

173.1
= 14.2 O.K.
11.3) CompressionFlangeBracingRequirement (AASHTOArticle6.10.10.2.4)
(a) Flexure(AASHTOArticle6.10.6.4.1)
To ensurethat anoncompositesteel girder hassufcient exural resistanceduring
construction, themoment capacityshouldbecalculatedconsideringlateral torsional
bucklingwith an unbraced length, L
b
(Figure10.77).
For asteel girder without longitudinal stiffenersand (2D
c
/t
w
) >
b
_
E/F
yc
, the
nominal exural resistanceis
M
n
=
_

_
1.3R
h
M
y
M
p
For L
b
L
p
C
b
R
b
R
h
M
y
_
1 0.5
_
L
b
Lp
LpLr
__
R
b
R
h
M
y
For L
p
< L
b
L
r
C
b
R
b
R
h
My
2
_
Lr
L
b
_
2
R
b
R
h
M
y
For L
b
> L
r
(10.83)
L
p
1.76r
t
_
E
F
yc
(10.84)
L
r
=
_
19.71I
yc
d
S
xc
E
F
yc
(10.85)
where
b
equals 4.64 for a member with a compression ange area less than the
tension angeareaand 5.76 for memberswith acompression angeareaequal to
or greater than thetension angearea; r
t
istheminimumradiusof gyration of the
compression ange of the steel section about the vertical axis; S
xc
is the section
modulusabout thehorizontal axisof thesection to thecompression ange(equal
to S
st
in Table 10.24); C
b
is the moment gradient correction factor; and R
b
is a
angestressreduction factor consideringlocal bucklingof aslender web(AASHTO
Article6.10.5.4.2).
c 1999by CRCPressLLC
C
b
= 1.75 1.95
_
P
l
P
h
_
+0.3
_
P
l
P
h
_
2
2.3 (10.86)
whereP
l
istheforcein thecompression angeat thebraced point with thelower
forcedueto thefactored loading, and P
h
istheforcein thecompression angeat
thebraced point with higher forcedueto thefactored loading. C
b
isconservatively
taken as1.0in thisexample.
r
t
=
_
I
yf
A
f
=
_
(18)
3
(1.0)/12
(18)(1.0)
= 5.20 in. (132 mm)
L
p
= 1.76r
t
_
E
F
yc
= 1.76(5.2)
_
29,000
50
= 220 in. < L
b
= 240 in.
L
r
=
_
19.71(486)(98.75)
2,898
29,000
50
= 435 in. (11,049 mm)
2D
c
t
w
= 173.1 <
b
_
E
f
c
= 4.64
_
29,000
20.95
= 172.6
Sincethesetwo valuesarevery close, takeR
b
=1.0(AASHTOArticle6.10.5.4.2).
M
y
= S
st
F
y
= 2,898(50) = 144,900 kips-in. = 12,075 kips-ft
..
.
L
p
= 220 in. < L
b
= 240 in. < L
r
= 435 in.
..
.
M
n
= (1.0)(1.0)(1.0)(12,075)
_
1 0.5
_
240 220
435 220
__
= 11,513 kips-ft < R
n
R
h
M
y
= 12,075 kips-ft
M
u
= 0.95(1.25)(4,260) = 5,059 kips-ft (6,859 kN-m)
<
f
M
n
+(1.0)(11,513)
= 11,513 kips-ft (15,609 kN-m) O.K.
(b) Shear (AASHTOArticle6.10.10.3)
Check thesection at therst intermediatetransversestiffener, 20 ft (6.10 m) from
theleft end of span 1. V
u
istaken conservatively fromthelocation of 0.1L
1
.
V
u
= 0.95(125)V
DC1
= 0.95(1.25)(107.5) = 120.9 kips (538 kN)
For an unstiffened web, V
n
=335.5kips(1,492kN); therefore, weobtain:

V
n
= (1.0)(335.5) = 335.5 kips > V
u
= 120.9 kips O.K.
c 1999by CRCPressLLC