review article a review on strengthening steel beams using...

Review ArticleA Review on Strengthening Steel Beams UsingFRP under Fatigue

Mohamed Kamruzzaman Mohd Zamin JumaatN H Ramli Sulong and A B M Saiful Islam

Department of Civil Engineering University of Malaya 50603 Kuala Lumpur Malaysia

Correspondence should be addressed to Mohd Zamin Jumaat zaminumedumy

Received 30 December 2013 Revised 9 July 2014 Accepted 9 July 2014 Published 27 August 2014

Academic Editor Hua-Peng Chen

Copyright copy 2014 Mohamed Kamruzzaman et al This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

In recent decades the application of fibre-reinforced polymer (FRP) composites for strengthening structural elements has becomean efficient option to meet the increased cyclic loads or repair due to corrosion or fatigue cracking Hence the objective of thisstudy is to explore the existing FRP reinforcing techniques to care for fatigue damaged structural steel elements This study coversthe surface treatment techniques adhesive curing and support conditions under cyclic loading including fatigue performancecrack propagation and failure modes with finite element (FE) simulation of the steel bridge girders and structural elements FRPstrengthening composites delay initial cracking reduce the crack growth rate extend the fatigue life and decrease the stiffness decaywith residual deflection Prestressed carbon fibre-reinforced polymer (CFRP) is the best strengthening option End anchorageprevents debonding of the CRRP strips at the beam ends by reducing the local interfacial shear and peel stresses Hybrid-jointnanoadhesive and carbon-flex can also be attractive for strengthening systems

1 Introduction

The collapse of structural elements due to fatigue is extremelyexpensive and may be catastrophic in terms of human lifeand damage The fatigue process can be defined as theaccumulation of damage at a localized region as a result ofcyclic loads which leads to the formation of a crack thateventually propagates through the material When a crackgrows to a size forwhich the net-section is inadequate to carrythe imposed load rapid fracture takes place Fatigue failure ofsteel girders is one of the most significant problems affectingthe remaining service life of existing steel structures likebridges The application of fibre-reinforced polymer (FRP)composites to increase the fatigue strength of damaged steelgirders is a promising technique that offers an attractivesubstitute to traditional approaches like steel plating

In recent decades the strengthening of the steel com-ponents of structures has become essential for structuralretrofits FRP strengthening seems to be an effective alter-native for such retrofitting A survey of European rail-way administrations covering nearly 220000 bridges across

Europe [1] has shown that almost 22 of the bridges areconstructed with metal Among them 28 of the metallicbridges are over 100 years old and 40 are between 50and 100 years old The federal highway administration ofthe US Department of Transportation 2005 [2] estimatedthat among the almost 200000 metallic highway bridgesin the US around 40 are either ldquostructurally deficientrdquoor ldquofunctionally obsoleterdquo Deterioration is mostly due toaccidental vehicular impacts cyclic loads Bridge structuresthat support moving trains are subjected to high stressescaused by extreme vibrations and dynamic deflections thatare far greater than ever before [2] Moreover the fatiguelife of steel structures could be extensively reduced due toearthquake loadings [3] The dynamic reaction of railwaybridges is also influenced by various factors including thetrain speed natural frequency of the bridge and the bridgeand carriage lengths [4] The I-35W Bridge over the Missis-sippi river collapsed on 1 August 2007 resulting in 13 deadand 145 injuries and damaged 111 vehicles The NationalTransportation Safety Board [5] cited on the collapse of theI-35W Bridge that a design fault improper maintenance and

Hindawi Publishing Corporatione Scientific World JournalVolume 2014 Article ID 702537 21 pageshttpdxdoiorg1011552014702537

2 The Scientific World Journal

additional weight because of increasing traffic volume andheavy traffic caused the fatigue failure The fatigue damagereported in Japan has been causedmainly by cyclic secondarystress or distortion-induced stress [6]

Fibre-reinforced polymer possesses outstanding advan-tages as a structural material including high strength anti-corrosion properties and high durability and is able torestore the lost capacity of damaged structures [7 8] FRPsheetsstrips are also effective in the strengthening of steelstructural elements to extend their fatigue lifetime and reducecrack propagation [9ndash11] if galvanic corrosion is preventedand sufficient bond is provided [12 13] Recent strengtheningprojects in the United States Switzerland the United King-dom and Japan showed that there was a great potential forusing carbonfibre-reinforced polymer (CFRP) to retrofit steelstructural elements [14ndash16] CFRP materials have been usedfor marine structures [17 18] and aerospace parts [19] andhave expanded to wood [20] masonry [21] steelconcretecomposite structures [22] and so on It is more frequentlyused for rehabilitation and strengthening of steel structuresthan other FRP materials due to its high strength In anycase CFRP is very tolerant to fatigue damage [23 24] Basalt-fibre-reinforced polymers (BFRP) have been increasinglyconsidered in civil infrastructures because of low cost andtheir excellent chemical and mechanical properties [25 26]

The Acton bridge in England [27] and the Ashlandbridge and 78385S092 bridge in the USA were retrofitted byapplying CFRP elements to the bottom of the girders thestress reduction on the original materials was observed toincrease the fatigue life [28] Several case studies of existingsteel railway bridges are presented in [29 30] Monitoringdata was used for dynamic response analysis and fatigue lifeevaluation It was observed that the stringers and the crossbeams run a high risk of fatigue damage The weakest pointin the reinforcing system of FRP elements to steel joints is thebond of the adhesive [31ndash33] The successful implementationof FRP composites of the strengthening systems is dependentupon the quality and integrity of the steel-composite jointand the effectiveness of the epoxy adhesive used [34] Thefatigue performance of the CFRP reinforcement details isnecessary to check if under realistic severe spectrum loadingthe adhesive joint performs better than the common fatiguesensitive details found on steel bridge girders in accordancewith Eurocode 3 categories 36lowast and 50lowast [35] or AmericanAssociation of State Highway and Transportation Officials(AASHTO) categories E1015840 and E [36]

The application of FRP composite for strengthening steelbridges and structural elements has become an efficientoption to meet the increased cyclic loads or repair due tocorrosion or fatigue cracking Therefore the objective of thisreview is to explore the potential FRP reinforcing techniquesto care for fatigue damaged steel structural elements Thispaper reviews the research related to CFRPsteel strength-ening techniques under fatigue Detailed knowledge andexplanation of the existing research concerning the fatiguebehaviour of HM HS UHM and prestressed CFRP and SW-BFRP strengthened steel structures are provided The studyalso covers the surface treatment techniques adhesive curingand support condition under cyclic loading including fatigue

performance crack propagation and failure modes with FEsimulation of the steel bridge girders and structural elementsFuture research gaps and recommendations are indicatedaccordingly

2 Surface Preparation and Treatment forSteel Beam Strengthening

21 Types of Notch To simulate the actual damage causedby corrosion and the expansion of fatigue cracks severalresearchers intentionally created notches of different geome-try in midspan or other positions on the tension flange of thebeams (Figure 1(a)) In addition the notch assists like a stressconcentrator [37] in the damage-sensitive regions [38 39] tocommence a vertical crack at the steel webThedifferent typesof notch can be categorized as follows

(a) rectangular notch on both edges (Figure 1(A))(b) U-shaped notch on both edges (Figure 1(B))(c) U-shaped notch through the whole tension flange

(Figure 1(C))(d) nonuniform notch (Figure 1(D))(e) uniform notch (Figure 1(E))

Usually the researchers cut a notch at midspan in thetension flange of the steel beams except Kim and Harries[37] To initiate the debonding of CFRP which was aimed atpropagating towards the right support they created a notchthrough the entire tension flange at a position 152mm to theright of midspan of the beams Jiao et al [40] welded the cutalong the tension flange soffit using the shielded metal arcwelding (SMAW) approach

The notch (C) through the whole tension flange issensitive to fatigue compared to the side notches (A) and(B) Accordingly notches (D) and (E) which go throughthe whole tension flange with part of the web in midspanare more sensitive to fatigue damage as the damage occurssuddenly

From the above categories of creating the notch it hasbeen revealed that when the notch spreads through theflange as well as the web of the steel beam the propagationof the cracks exists in the notch line of the flange andthe web Brittle fracture can happen in the case of fatigueThis is injurious for the structural element as no warningis given before failure In addition notches only createdin the flange can expect a retarding fracture with priorwarning However if the notch is given in the whole tensionflange there is also the possibility of brittle fracture underheavy repeated load Therefore for observing the properfatigue damage a rectangular and U-shaped notch on bothedges at midspan may be competently incorporated for thedevelopment of a standardized test Figure 1(b) illustratesthe stress characteristics of a strengthened steel beam undercyclic loading for incorporation of different notchesThe S-Ncurves show that the uniform notch comprising whole flangeand web has the least stress compared to others This clearlydemonstrates more fatigue life when the side rectangularnotch is incorporated in the flange

The Scientific World Journal 3

CFRPAdhesive

Flange

Web

CFRPFlange

Web09 mm

127 mm 127 mmNotch

(A) Rectangular edge notch (Cut front view Cut section

Bottom flange

Notch

218 mm 218 mm

8 mm

(B)U-shaped edge notch (

Notch

152 mm

FlangeWeb

32 mm radius

t f=71

mm

Centre line of midspan

FlangeWeb

15mm

5mm

1mm

5mm

Notch

tf

(D) Nonuniform notch (Ghafoori et al 2012)

Flange

Web

Notch

(E) Uniform notch (Jiao et al 2012)

(Tavakkolizadeh and Saadatmanesh 2003)

(Wu et al 2012)

(C)U-shaped notch (Kim and Harries 2011)

(a)

Figure 1 Continued


10

100

1000

10000 100000 1000000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

Rectangular notch in both edges (Tavakkolizadeh and Saadatmanesh 2003) U-shaped notch through the whole tension flange (Kim et al 2011)Uniform notch (Jiao et al 2012)

(b)

Figure 1 (a) Types of notch in steel beam (b) Stress behaviour in fatigue for different notch categories [14 37 40]

Cover plate

Stiffener

Steel plate SW-BFRP HM-CFRP HS-CFRP HS-CFRP HS-CFRP

(i) (ii) (iii) (iv) (v) (vi) (vii)

Figure 2 Strengthening technique with plate configuration [38]

22 Welding Cover Plate and Stiffener To provide the equiv-alent of a concrete slab that typically exists in bridges for pre-venting compression flange buckling Wu et al [38] attacheda steel cover plate with welding on the outside surface of thetop flange of the steel beams (Figure 2) They also weldedweb stiffeners on both sides of the web at the loading andsupporting points Web stiffeners assist in preventing webcrippling at the midspan section [41] In order to providelateral stability of the steel beams Kim and Brunell [42]used stiffeners that were welded at the supporting pointsThe adhesively bonded steel stiffeners to the flanges andwebs on both sides of the beam extensively retarded localbuckling of the steel beams [43] Siddique and El Damatty[44] showed that the use of glass fibre-reinforced polymer(GFRP) composites enhances local buckling behaviour ofwide flange steel beams which is effective especially forthe case of slender beams The addition of GFRP platesto the compression flange of a steel beam increases boththe load carrying capacity and the deflection at failure Theimprovement in the load capacity is independent of the webdimensions of the beams for both plastic and slender beamsThe study ignored the use of a stiffenerThemode of failure ofthe retrofitted slender beams ranges from elastic buckling ofthe system to GFRP rupture when the thickness of the GFRP

is varied from 635mm to 190mm As the GFRP thicknessis significant in their study it could well improve the localbucking

23 Prevention of Galvanic Corrosion Although CFRP is anoncorrosive substance when carbon fibres are in contactwith steel they can form a galvanic cell To increase thefatigue strength of bridge girders and long-term durabilityof CFRP reinforcement in a steel structure the preventionof galvanic corrosion is necessary Furthermore to ruleout the galvanic corrosion the flow of corrosion needs tobe prevented This may be accomplished by insulating thedifferent metals from one another or through preventinga continuous link of electrolytic solution between the twoby coating with a water resistant sealant [45] It is obvi-ous that if the two different metals are not in contactgalvanic corrosion cannot occur [46] Tavakkolizadeh andSaadatmanesh [47] examined the galvanic corrosion betweencarbon and steel for various thicknesses of adhesive coatingin different electrolytes such as seawater and deicing saltsolution The thin coating effect of adhesive (025mm) wasfound to be substantial as was the sizing applied to the fibresFurthermore a thicker adhesive between the surfaces of the


CFRP and steel was observed to suggestively slow the rate ofcorrosion of the steel

Mitigating galvanic corrosion of the CFRP-steel compos-ite can be achieved by the selection of an adhesive with goodquality isolation properties [15] or by using a thicker epoxywater resistant sealant or nonconductive layer plus a sealantor by bonding a GFRP sheet before applying the CFRP layeronto the steel surface [17 48ndash50] Hollaway and Cadei [51]installed a polyester drape veil to provide insulation betweenthe steel and the carbon fibre for preventing direct contactbetween them Fibreglass or an epoxy film was consideredto provide effective insulation In addition a monitoringprogramme was initiated to identify the cathodic sites so thatgalvanic corrosion damage could be mitigated or stopped[34]

24 Surface Treatment The reliability of the joint is highlydependent upon the surface treatment processes for bondingthe fibre-reinforced composite to the steel structural elements[52] The surface preparation and the strength of the appliedCFRP overlay can significantly affect the fatigue performance[53] For assessing the effect of the CFRP strengtheningtechnique on the fatigue strength Jiao et al [40] used agrinder to remove the corrosion as well as to level the weldarea on each steel beam soffit before applying the adhesive Toobtain a clean rough and chemically active surfaceWu et al[38] treated the surface of the tension flange using a grindingwheel to reinforce with CFRP for the fatigue testThe surfacesof the tension flange and composite plates were then washedwith acetone

Tavakkolizadeh and Saadatmanesh [14] used a sandblastermeticulously by number 50 glass bids andwashedwithsaline solution just prior to the application of the compositesheet to prevent oxidation The study by Teng et al [54]showed that sand blasting was the most effective surfacetreatment Prior to bonding the CFRP strips to the beamsKim and Harries [37] used a 1500 sfpm (surface feet perminute) belt sander and a 40 grit zirconia alumina belt Thisensured a sound slightly striated surface to bond the CFRPstrips The adhesive layer thickness was approximately 1mm

Deng and Lee [55 56] found that the tips of the FRPplates must be finished smoothly using sandpaper beforethe attachment of the plate to the steel beams HoweverSchnerch et al [57] disagreed with Choudhury [58] as theycontended that preparing the surface with a hand grinderfollowed by sandpapering reduces the bonding ability of thesurface However a chemically active steel surface that is freefrom contaminants is essential to enhance the chemical bondbetween the adhesive and the metallic surface Brushingultrasonic or vapour degreasing systems are claimed to bethe most efficient to remove oil and other potential surfacecontamination especially when adequate solvents are used[59] Contamination may then be removed using the excesssolvent rather than simply redepositing it on the steel surfaceas the solvent evaporates

The most efficient means of achieving a high-energysurface of the steel is by grit blasting [51] Grit blastingwith angular grit removes the inactive oxide and hydroxide

deposits by cutting and deformation of the base materialThegrit size also affects the surface profile of the steel Harris andBeevers [60] stated that finer particles created a smoothersurface than coarser grit particles and smoother surfacesexhibited higher adhesive-steel surface bonding In additionthe surface profile of the steel was not influenced on the long-term durability [34] After grit blasting solvents may be usedto wash and clean the steel surface without resulting in poorbonding [61 62]

3 Fatigue Strengthening Techniques forSteel Beams

The modulus of elasticity tensile strength shape and con-figuration of FRP composites of an adhesively bonded jointplay an important role in respect of the fatigue strengthand lifetime of reinforced steel beams and bridge girdersA number of researchers have investigated reinforced steelbeams with different FRP strengthening techniques andcompared their fatigue performance A summary of the rein-forcing technique of steel structures using fibre-reinforcedcomposites is provided in Table 1

Furthermore Wu et al [38] investigated eight artificiallydamaged H350 times 175 steel beams including one unstrength-ened and seven strengthened with welded steel SW-BFRPhigh modulus CFRP (HM-CFRP) and high strength CFRP(HS-CFRP) plates using Sikadur-30 Normal epoxy adhesiveThe plate configuration of the strengthening technique usedby Wu et al [38] is shown in Figure 2 The elastic modulusultimate strain tensile strength and shear strength of theepoxy were 2627GPa 15 317MPa and 144MPa respec-tively An anchorage system was applied at below the pointload and at the end of the fibre-reinforced composite platesThe HM-CFRP has the most excellent strengthening effectsand SW-BFRP is the best strengthening material on the basisof the cost-performance ratio [38]

Basalt-FRP (BFRP) composites show synthetical advan-tages in structural strengthening seismic rehabilitation andserving as new structural materials [26 65] However therelatively low modulus of BFRP may not satisfy the stiffnessrequirement of some structures Therefore to obtain higherperformance steel-wire- (SW-) BFRP can be made fromhybridization of BFRP with steel wires (SW) or CFRPs [2666]

A 210 ft times 26 ft three continuous span rolled steel bridgegirder in Guthrie County Iowa on state highway 141 wasstrengthened using externally posttensioning CFRP rods[67] The anchorage systems were bolted to the webs of thesteel girdersThe proposed prestressed unbounded reinforce-ment (PUR) system [64] can be applied as an alternative toadhesively bonded FRP reinforcement mainly when there isconcern about the effects of water moisture high ambienttemperatures and high cycle loading on the glue betweenthe FRP and the steel Vatandoost used 14 15 35 0and 37 prestressed CFRP plates to investigate the fatiguebehaviour of five W310 times 74 steel beams Vatandoost [68] inwhich the 14 15 and 35 prestressed CFRP plates werebonded to the inner side of the bottom tension flange and


Table 1 Strengthening techniques

Specimensreference Material-dimensions and properties Figures

S127x45 steel beamsreinforced with CFRPsheet [14]

Steel beam 122m long 119891119910= 3364MPa 3309MPa

119864 = 1944GPaCFRP 300mm times 76mm times 127mm 119864 = 144GPa119891119906= 2137MPa

Epoxy 1 1 mixer of resin (bisphenol A based) andhardener (polyethylene polyamine)

Notch at midspan

Adhesive

One-layer CFRP sheet

S127x45

127x76UB13 steel beamsstrengthened usingCFRP plate [55]

Steel beam 1200mm long 119864 = 205GPa 119891119906= 275MPa

FRP 3mm thick and 400mm longEpoxy (Sikadur 31 Normal) 119864 = 8GPa119904hear modulus = 26GPa 119891

119906= 297MPa and 03mm

thick

One-layer CFRP plateAdhesive7x

W150x18 steel beamsrepaired with CFRPstrips [37]

Steel beam span length = 1830 mm 119864 = 200GPa 119891119910=

393MPaCFRP 50 times 14mm 119864 = 155GPa 119891

119906= 28GPa

Epoxy 119864 = 45GPa 119891119906= 25MPa 1mm thick One-layer CFRP strip

Adhesive Notchx18

Grade 400 150UB14notched steel beamsretrofitted with weldingCFRP composites [40]

Steel beam L1400 timesW75 timesH150mm 119864 = 2074GPa119891119910= 4116MPa 119891

119906= 5413MPa

CFRP(1) Sika CarboDur M1214 pultruded plates-14mmthick 119864 = 210GPa Ft = 24GPa(2) SikaWrap Hex-230C woven sheets-013mm for eachply 119864 = 230GPa 119891

119906= 345GPa

Epoxy(1) Sikadur-330(2) AralditeR 420

CarDur M1214 plate (1 layer)

SikaWrap Hex-230C sheet (4 layers)

S355J0 steel beamsbonded with 20prestressed CFRP plates[63]

Steel beam 1100 times 120 times 65mm 119864 = 210GPa119891119910= 355MPa

CFRP 910 times 50 times 12mm (S512) 119864 = 165GPa 119891119906=

310GPaPrestressing level 20 of the ultimate CFRP strength =632MPaAraldite 2015 adhesive 119864 = 175GPa

Prestressed CFRP plate

Notch

Adhesive

S355J0 (ST 52-3)

S355J0 (ST 52-3) steelI-beams strengthenedusing 30 (a)prestressed unboundedand (b) bonded CFRPplates [64]

Steel beam 1100 times 120 times 65mm 119864 = 210GPa 119891119910=

355MPaCFRP 910 times 50 times 12mm (S512) 119864 = 160GPa 119891

119906=

310GPaPrestressing level 30 of the ultimate CFRP strengthAraldite 2015 adhesive 119864 = 175GPa


Notch

Anchorage

S355J0 (ST 52-3)

(a) PUR system


Notch

AdhesiveAnchorage

S355J0 (ST 52-3)

(b) PBR system

the 0 and 37 prestressed plates were attached to the coverplate A prestress FRP composite patch is strongly suggestedto maximize the effectiveness of the adhesively bonded patchon the steel element [69] and fatigue strengthening [70] TheFRP prestressing system developed at Empa was used forthe CFRP strips for applying a direct tensile force againstan external reaction steel frame by jacking Vatandoost and

others [68 71ndash73] discussed more details concerning theprestressing procedure

Recently the carbon-flex that is carbon fibre hybrid-polymeric matrix composite (CHMC) strengthening tech-nique was developed by Zhou and Attard [74] which is acarbon fibre-based composite manufactured using the latesthybrid-matrix technique involving amino-based polymeric


composites to provide necessary damping and high strengthsustainability of the carbon fibrous element Recently Zhouand others [74ndash76] indicated the enormous potential ofcarbon-flex as a strengthening substance to subsequentlyprevent higher damage or catastrophic failure of structures

The majority of fatigue problems arise from poor detail-ing or careless fabrication rather than inaccurate materialsselection [77] Schnerch and others [15 52] reported that thebonding mechanisms of FRP strengthened steel structuresare different than concrete structures In addition high bond-ing stresses occur in steel structures tomeet the strengtheningrequirements [34] Any violation of fabrication tolerances canunpredictably change the fatigue behaviour and lead to a veryscattered fatigue life [73]

4 Adhesive Curing

If a bridge or long span structure is retrofitted with CFRPstrips it is generally not economic to close it to trafficduring the adhesive curing time which can take up to 48hours During this time the epoxy adhesive is subject torepeated loading from the traffic The Concrete Society [78]recommended that the change in the epoxy properties causedby the repeated load during the curing time is expected tobe small perhaps a 10 decrease in the strength of the fullycured structural elements Nikouka et al [79] studied theimprovement in the strength and stiffness of strengthenedsteel beams with CFRP subjected to repeated loading duringthe early age curing of the epoxy adhesive Five pairs of 127 times76UB13 type steel beams each 12m long were strengthenedwith a 098m long single K13710 ultra-high modulus CFRPplate attached to the bottom tension flange A cyclic load wasapplied to the five specimens with 025Hz frequency and wascontinued for up to around 48 hoursThe study reported thatduring the curing of the epoxy the cyclic loading would affectthe final stiffness and failure load of the strengthened beamwhen the highest cyclic loadwas larger than 42 kNMoreoverthe bond would fail to develop if the shear deformationin the epoxy layer during the cure is too large They alsorecommended that it was prudent to limit the shear stress inthe epoxy to a maximum of 1MPa

Bourban et al [80] indicated a clear advantage fromthe epoxy adhesive curing at high temperatures (about93∘C) during the initial cure (10ndash20 minutes) The resultingbond is stronger tougher and more durable when subjectto unfavourable environments [81] With the intention ofretrofitting steel bridges open to traffic during the adhesivecuring period Moy [27] investigated the effect of repeatedloading on the curing of the epoxy The results confirmed aprogressive stiffness increase of the reinforced component asthe epoxy cured Furthermore the beams subject to higherloads during curing did not develop the full bond of the epoxyadhesive The tests performed showed that cyclic loadingat higher load levels reduced the ultimate capacity of thestrengthened beams [41] In addition the vibration of thetraffic during the curing of the adhesive causes a progressivereduction in the fatigue lifetime with increasing strain level[82 83]

Zhang et al [84] proposed an innovative method involv-ing preimpregnation (prepreg) advanced composites and acompatible epoxy film for retrofitting steel railway bridgesopen to traffic during the curing period of the epoxy adhesiveThe strengthening system was made from unidirectionalHM- and UHM-CFRP preimpregnations that were cured onsite under vacuum assisted pressure Two cure temperatureswere used 65∘C for 16 hours and 80∘C for 4 hours A GFRPprepreg layer was placed between the CFRP patch and thesteel element The beams were initially induced by vibrationforces and then loaded to failure From the experimentalresults it was observed that despite slight damage at theadhesive level the proposed technique prevented severebrittle failure of the composite beam

5 Support Condition and Instrumentation

Different support conditions have been adopted by differentresearchers for the fatigue test programmes of steel beamsas shown in Figure 3 Deng and Lee [55] tested the fatiguestrength of nine reinforced steel girders by a servo-hydraulicDennison testingmachine using a three-point bending setupas a simply supported beam (Figure 3(a)) The specimenswere supported on two rollers but were restrained fromany sideways movement The loading block had two steelplates each with a counter seat and a roller in betweenDeflections were measured at three locations by means ofpotentiometers Five 2mm and two 5mm long strain gaugeswere used to investigate the crack initiation as well as theeffect of crack growth on the stress field in the girder Allthe data were recorded using a data logger Studies on thefatigue of double sided reinforcement subjected to tensionand full-scale bridge girders retrofitted with CFRP platesunder three-point bending were conducted at the Universityof Delaware [85] Kim and Harries [37] used a neoprenerubber pad between the support and beam to reduce theconcentration of stress as shown in Figure 3(b) In all casesthe CFRP plates remained fully bonded to the steel elementThe results suggested excellent fatigue behaviour of thereinforced elements

Wu et al [38] tested the strengthened H350 times 175 steelbeams under constant amplitude cyclic load using 4Hz fre-quency as a simply supported mode and four-point bendingas shown in Figure 3(c) The load was measured by theloading cell of aMTS system To prevent anymovement of thespecimen during the test Tavakkolizadeh and Saadatmanesh[14] used tie down brackets to the roller supportsThe loadingblocks were designed using a counter seat for the compres-sion flange in order to prevent their movement during theexperiments The loading setup is shown in Figure 3(d) Thespecimenswere tested using various constant stresses rangingbetween 69 and 379MPa (119877 = 01) and a frequency of 5and 10Hz Vertical displacements can be measured by linearvariable displacement transducers (LVDT) with a range ofplusmn50mm [38] and plusmn75mm [14]

Jiao et al [40] conducted fatigue tests under load controlwith 7Hz on strengthened steel beamswith a 4-point bendingrig using a MTS-810 testing machine which contained the


Roller

Hydraulic jack

Steel plate

170

CFRP plate127x76UB13

Adhesive

Support beam

Steel plate300 350180 5050

Roller support Roller support

Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm

Strain gauge on web above notch

763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages

Load cellSteel cover plate

188

70

70

Notch H350 times 175

(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell

Roller support with steel bracket


(d) [14]

Figure 3 Continued


395 mm

CFRP 150UB14Adhesive

Top supporting frame

Pin connected wheelPin connected wheel

Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel

(200 times 100times 5 RHS)

Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]

Figure 3 Support condition with fatigue test setup

top supporting frame and bottom loading beam as shownin Figure 3(e) Two 12mm thick steel plates were welded tothe midspan of the top supporting frame and the bottomloading beam Four pin connected wheels were employed atthe supporting and loading points that could freely rotateduring the fatigue tests Two screw-fixed stoppers wereused on both sides of the bending rig to prevent the testspecimen fromchanging position during the experimentThebottom loading rig was designed using a three-pin system toensure that the load was distributed between the two loadingpoints Using SHOWA strip strain gauges the ultimate loaddisplacement number of cycles and corresponding strains ofeach cycle were recorded using the National Instrument NI9237 Compact Data Acquisition system

To fatigue test the reinforced metallic beams using pre-stressed FRP Ghafoori et al [64] used a pulsator P960 oilhydraulic test machine with a four-point bending setup Thelubricated rollers of 5 cm diameter at the supports and asteel plate were employed between the beam and rollersto distribute the load properly (Figure 3(f)) Ghafoori et al[63] used a 3D ICS (image correlation system) to measure

the crack length and the corresponding strain deformation atthe crack tip area The measurement window of the ICS wasset at 65mm times 65mm The calibration details and the use ofthe ICS can be found in [87ndash89] The field signature method(FSM) is also effective for detecting andmonitoring cracks onsteel structures [90]

6 Finite Element (FE) Simulation

The finite element method (FEM) is an acceptable approachfor analysing structures using software In practice the FEsimulation is developed to validate the fatigue strength of theexperimental or analytical results

Based on the surface crack widening energy release rate[91] using an elementary material strength theory [92] andGlowast-integral [93] an analytical approach was introduced byGhafoori and Motavalli [87] to estimate the stress intensityfactors (SIF) of a cracked steel I-beam The fatigue rehabili-tation of steel structures is usually expected to decrease thevalue of SIF at the tip of the crack and as a result enhance


Crack tip

Crack length

Symmetric plane along the webSymmetric plane at midspan

X

Y

Z

(a)

Anchorage

LoadingMesh refinement

zone

CFRP plate

X

Y

Z

(b)

Figure 4 (a) A geometrical model using ABAQUS in the FE analysis and (b) the mesh refinement around the loading anchorage and crackzones [64]

the postcrack fatigue life [94] Ghafoori et al [64] proposedan analytical method using the experimental test data (theexternal bending moment the length of the crack and thecorresponding strain imposed on the CFRP strip under thecracked segment) and produced the SIFThey used ABAQUSsoftware (version 68) to analyse the FE model of the steelbeams to validate the results The geometrical model andmore mesh refinement around the loading anchorage andcracked sections are shown in Figure 4 The method wasdeveloped to assess the sufficient level of the CFRP prestress-ing to arrest the fatigue crack growth (FCG) Moreover themethod was used to study different active semiactive andpassive crack modes with a loaded reinforced beam Severalfactors have been considered including crack propagationexcitation frequency and structural damping on the life of theFCG [95]

Using the concept of fracture the fatigue crack propa-gation (FCP) model was proposed by Xiulin and Hirt [96]This was extended to the FCP of a cracked metallic elementretrofitted with adhesively bonded composite patches in thestudy of Wang and Nussbaumer [97]

According to the Paris-Erdogan crack growth law [98]a linear elastic fracture mechanics (LEFM) model wasemployed to predict the effects of peening treatments onthe fatigue performance of welded steel structures [99] andto confirm the effectiveness of the prestressed CFRP strips[68] Ghafoori et al [63] introduced a methodology for adamaged beam with a specific crack length that is subjectedto a certain cyclic load based on the fracture mechanics(FM) theory to estimate the adequate prestressing level bywhich the crack propagation is detained Some strengthenedbeams were tested under various cyclic loading ranges andthe experimental results showed excellent agreement with thedeveloped fracture model

In fracture mechanics based on progressive damagemodellingmethods to predict the fatigue life the rate of crackgrowth is related to the SIF [98] or strain energy release rate[100ndash103] In the case of adhesively bonded joints a damageshift parameter was proposed to account for the effectsof accelerative interaction [102 103] Important interactioneffects were considered where crack growth acceleration was

linkedwithmean load changes However continuumdamagemechanics (CDM) models were developed by Lemaitre andDesmorat [104] and modified for the damage formation ofmicrocracks by [105 106] CDM models have been usedin a damage evolution law for modelling both precrackingdamage evolution and crack growth for constant and variableamplitude fatigue [107] In the case of bonded joints Wahabet al [108] compared both the FM and the DM methodsto predict the fatigue strength of adhesively bonded CFRPdouble lap joints They verified that the developed CDMapproach compared favourably with a FM method for con-stant amplitude fatigue (CAF)The FM andDMbased fatiguelife prediction of bonded single lap joints (SLJs) subjected todifferent types of variable amplitude fatigue (VAF) loadingwas analysed by Shenoy et al [109]

Kim and Harries [37] developed a three-dimensional(3D) nonlinear finite elementmodel for predicting the fatiguestrength of notched steel beams using ANSYS softwareThe steel section was modelled using 3D structural solidelements (SOLID45) and a linear stress-strain relationshipwas developed for the CFRP A nonlinear interface element(COMBIN39) with two nodes was applied for modellingthe behaviour of the steel-CFRP interface For the elementwhose initial relative distance is zero a bilinear bond-sliprelationship was created for them The study used the strainlife method and the concept of Henryrsquos damage theory [110]for the fatigue life prediction of steel beams The strain lifeapproach is mainly relevant to a member representing sig-nificant plasticity induced by hysteretic loadsThe theoreticalbackground of this approach is discussed by Bannantine etal [111] The deflection behaviour of unstrengthened andstrengthened beams is shown in Figure 5(a) Furthermorea typical S-N curve of strengthened steel beam obtained isshown in Figure 5(b) which was compared with category Ein the AISC in the study by [112] The notch provided forthe stress concentrating effect is essentially equivalent to aCategory E detail Apart from this Youssef [113] developeda model for predicting the linear and nonlinear behaviourincluding the deflection at midspan strains of the steel andFRP failure mechanism and failure load of rehabilitatedsteel beams The model was founded on the solution of


0

50

100

150

200

0 5 10 15 20

Load

(kN

)

Vertical deflection (mm)

Unstrengthened (Exp)Unstrengthened (FEA)

Strengthened (Exp)Strengthened (FEA)

(a) Load deflection relationship in experimental and numerical studies(static) [86]

10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA

(b) S-N curve in experimental and numerical studies [37]

Figure 5 Comparison of experimental and simulation outputs of steel beam

the differential equations governing the behaviour of astrengthened steel beam which includes representation ofthe shear and peel behaviour of the epoxy adhesive Tovalidate the predictions of the model a W-shaped steel beamstrengthened using GFRP sheets was experimentally testedand excellent agreement was found between these results

Zhou et al [114] adopted the micromechanics basedfracture model and the cyclic void growth model (CVGM)for estimating extremely low cycle fatigue (ELCF) fractureof the column-to-beam connections during earthquakesThe model was verified by the experimental results of ninefull-scale connection tests In addition the refined finiteelement model was used to simulate the cyclic behaviourof the connection tests and the CVGM fracture indexwas calculated using the stress and strain time historiesThe number of cycles and the cumulative deformations toELCF fracture predicted by CVGM agreed well with theexperimental results The existing methodology also showedreasonable good accuracy for predicting the ELCF fracture ofcolumn-to-beam connections under inelastic cyclic loadings

Pipinato et al [115] used a LEFMmethod in a probabilistic[116] context to assess the fatigue reliability of steel bridgegirders in the presence of seismic loading This methodcould enable a better understanding of progressive damagephenomena due to fatigue problems and could give somenew insights to increase the remaining fatigue strength ofa large number of steel bridges in seismic regions Colombi[117] developed a suitable plasticity based [118 119] crackretardingmodel as an extension of the well-knownNewmanrsquosmodel [120 121] to estimate the reduction of crack openingdisplacement along with the magnification of the crackgrowth retardation of the reinforced notched steel plates

From the abovementioned literature it is revealed thatthe simulation in the finite element method can be a vitaltool to assist in strengthening beam analysis under fatigueThis is because it eventually decreases the experimental

cost in finance and time Good validation of the simulationwith practical experiments ensures the advantages of thestrengthening techniques However the characteristics ofstrengthened steel beamsunder fatiguewithout using notchesare still an interesting area to be exploredThis interest can beaddressed by FE simulation in a consistent manner

7 Fatigue Performance ofReinforced Steel Beams

A steel structure subjected to repeated load may eventuallyexperience significant fatigue damage during its life Anumber of researchers concentrated on fatigue strength andfatigue lifetime prediction of reinforced steel beams andbridge girders This indicates that there is a need to enhancethe fatigue strength and prolong the fatigue lifetime of steelstructures with adhesively bonded metalFRP strengtheningtechniques The fatigue behaviour of reinforced steel beamsusing nonprestressed and prestressed FRP composite is illus-trated below

71 Reinforced with Nonprestressed Polymer Composite Holl-away andHead [122] indicated that unidirectional continuousfibre polymer composites which essentially behave linearlyup to failure level when loaded parallel to the longitudinalfibres usually have good fatigue properties Jiao et al [40]compared the behaviour of notched steel beams using thewelding method and retrofitted with CFRP plates and sheetsrespectively under flexural cyclic loads In addition twodifferent epoxy adhesives that is Sikadur-330 and Araldite420 were used in this test The observations of the fatiguestrength of the specimens reinforced with CFRP compositeswere extensively longer than that of specimens repairedwith the welding method alone It was observed that thestrengthening systemwith one layer of CFRP plate adhesivelybonded could extend the fatigue strength of steel beams


Retrofitted (Tavakkolizadeh and Saadatmanesh 2003)

10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)

Number of cycles (times104)

Unretrofitted (Tavakkolizadeh and Saadatmanesh 2003)Strengthened with CFRP plate (Jiao et al 2012)Strengthened with CFRP sheet (Jiao et al 2012)Welded only (Jiao et al 2012)

Figure 6 S-N curve for steel beam strengthened with nonpre-stressed CFRP [14 40]

about seven times compared to the beams only repairedwith the welding method In addition the fatigue strengthwas extended about three times for beams strengthenedwith four layers of CFRP woven sheets and no significantvariation in fatigue strength could be observed for specimensstrengthened using the epoxy adhesives of Araldite 420 andSikadur-330 Mean S-N curves were obtained based on thetest data (Figure 6) which can be used for predicting thefatigue strength of steel beams strengthened with similarCFRP composite materials

Tavakkolizadeh and Saadatmanesh [14] demonstratedusing the S-N curve that the application of CFRP stripscould increase the fatigue lifetime of the structural elementsmore than three times (Figure 6) The design S-N curve forunreinforced and reinforced cut specimens was NS354 =122 times 10

13 and NS396 = 384 times 1014 respectivelyThe slope of the S-N curves for the specimens (retrofitted

and unretrofitted) in a log-log space was slightly smaller thanthe slope of the AASHTO design curves They also observedthat the CFRP patch not only decreased the crack growth ratebut also was able to carry a few extra cycles even after thetension flange had completely cracked especially under lowerstress ranges (Figure 7)

Deng and Lee [55] reported the results of an experimentalprogrammeon small-scale steel beams reinforced by applyingCFRP strips From the tests results an S-N curve wasobtainedThe fatigue limit that is threshold of the S-N curvewas about 30 of the ultimate static failure stress whichvalidates the fatigue limit recommended by theCIRIADesignGuidance [123] To assess the fatigue bond resistance of a steelbridge girder reinforced with CFRP strips Miller et al [85]conducted two test programmes First they subjected sevensmall-scale doubly reinforced specimens to cyclic loads at827MPa stress for 255 million cycles All CFRP strips werefound to remain fully bonded to the steel element withoutdeterioration based on the strain data taken before and after

Stress range = 207 MPa

0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles

FailureUnretrofittedRetrofitted

+71

97

+0

Figure 7 Fatigue crack growth curve for unreinforced and rein-forced steel beams [14]

the cyclic loading Subsequently two full-scale steel bridgegirders retrofitted with CFRP were fatigued for ten millioncycles at a constant stress range of 34MPa Throughout thetests the CFRP strips were inspected and monitored fordebonding but none was detectedTherefore the retrofittingtechnique was regarded as having good fatigue resistance

Abed [124] investigated the effects of temperature onthe adhesively bonded steel beams reinforced with CFRPcomposites The adhesive materials showed a significantreduction in the fatigue life and failure load of the strength-ened structures as the temperature reached the adhesive glasstransition temperature (119879

119892) Furthermore Keller and Scholl-

mayer [125] experimentally and numerically investigated thethrough-thickness performance of adhesively bonded FRPbridge decks and steel girders They found that no stiffnessdegradation occurred for cyclic loading of up to 10 millioncycles

Wu et al [38] investigated the fatigue behaviour ofstrengthened artificially notched steel beams including theeffects of the configuration and the number of layers ofHS-CFRP strips the interface treatment of the SW-BFRPcomposites and the type of materials Compared to thetraditional welded steel-plate approach the experimentalresults showed that the use of a fibre-reinforced compositestrip could not only delay crack initiation decrease the crackgrowth rate and prolong the fatigue life but also reduce theresidual deflection and stiffness decay The rough surface ofthe SW-BFRP could extend the fatigue strength of steel beamsmore effectively than using SW-BFRP with a smooth surfaceThey also used high modulus CFRP strips as a reinforcedmaterial HM-CFRP demonstrated the best strengtheningperformance the fatigue strength of the steel beams couldbe improved significantly by increasing the number of layersof the strengthening material Figure 8 presents the crackexpansion curves for four-layer and one-layer HS-CFRPWhen the number of layers was increased from one to fourthe crack initiation life and the fatigue lives were considerably


0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)

One layerFour layers

4 layers1 layer


Figure 8 Effect of layers of strengthening material [38]

enhanced (Figure 8) The plate configuration influences thefatigue strength

72 Reinforced with Prestressed Polymer Composite Althoughadhesively bonded FRP flexural strengthening techniqueshave been found to be an efficient approach to improvethe lifetime of fatigued steel structures there are relativelyfew studies that have applied prestressed CFRP strips tostrengthen against cyclic loadings Ghafoori et al [64]developed a prestressed unbounded reinforcement (PUR)method and compared the effectiveness and feasibility ofthe approach with the prestressed bounded reinforcement(PBR) method It could be used on heritage and historicalstructures where reversibility is important The experimentaltest results for the strengthened beams using the PBRmethodshowed a local strain concentration on the CFRP strip underthe cracked section while the PUR method had a uniformstrain distribution along the CFRP strip In addition thefatigue performance of the unbonded reinforcement systemwas better at a high prestressing level of the CFRP without asubstantial reduction in ductility

Ghafoori et al [63] studied the behaviour of notched steelbeams reinforced using prestressed and without prestressedbonded CFRP plates under cyclic loading The experimentalresults showed that the fatigue strength of a beam reinforcedusing the prestressed CFRP plate increased more than fivetimes that of an identical beam reinforced using nonpre-stressed CFRP plate (Figure 9) Both specimens were inducedby a cyclic load range of 9ndash90 kN

Vatandoost [68] drew deflection range curves for W310times 74 steel beam specimens strengthened with 14 15 350 and 37 prestressed CFRP plates The deflection rangeversus the number of cycle curves was drawn for the last45000 cycles Looking at Figure 10 it can be seen that (1) thehighest deflection range belongs to the control beam whilethe lowest deflection range belongs to specimen 37-C-Mindicating the highest stiffness increase for that specimen(2) The deflection ranges are dramatically increased at theend of the fatigue life (3) The lower deflection range for thespecimens with CFRP strips on the cover plates confirms

0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles

NonprestressedPrestressed

Figure 9 FCG curves for strengthened with nonprestressed andprestressed CFRP plates [63]

1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M

Figure 10 Deflection versus number of cycle curves for strength-ened steel beamusing 14 15 35 0 and 37prestressedCFRPplates [68]

the influence of the CFRP strip location on the specimenstiffness In Figure 10 ldquoFrdquo indicates that the strips are bondedto the inner side of the flange ldquoCrdquo indicates that the stripsare attached to the cover plate and ldquoS and Mrdquo indicate theCFRP strip with a standard modulus andmoderate modulusrespectively The beams with reinforcement located on thecover plates showed a greater fatigue life improvement thanthose with reinforcement located on the inner side of theflange [68]

73 Fatigue in Hybrid-Joint and Nanoadhesive In recentyears fatigue in hybrid adhesive joints which combinea traditional mechanical joint and a layer of adhesive(boltedbonded weldedbonded and rivetadhesive) hasattracted a considerable amount of researchers This is due to


Far Bottom Near

(a)

(i) (ii) (iii)

Steel beam fractured

Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)

Figure 11 Failure modes of retrofitted beams using FRP plates (a) [14] (b) [38] (c) crack propagation and brittle web fracture [37] and (d)end debonding [55]

their better fatigue performance compared to only mechani-cal joints or only bonded joints [126ndash130] Furthermore theuse of nanoadhesives (carbon nanotubes alumina nanoparti-cles and quartz nanoparticles) is a new field of application tobonded joints and has the potential to improve their fatigueperformance [127 131]

8 Failure Modes of Reinforced Steel Beams

The deterioration of the steel bridge structural capacity overtimemay be due to corrosion impact damage andor fatiguecracking [32 51 132]The crack propagation and failuremodeof FRP-strengthened steel structural techniques are normallydifferent than concrete-FRP techniques [15] Tavakkolizadehand Saadatmanesh [14] studied the performance of steel

I-beams with an edged notch in the tension flange andreinforced with a bonded CFRP patch under cyclic loadingThey demonstrated that cracking started from the tip ofone of the notches in the reinforced specimens and movedtowards the web (Figure 11(a)) In Figure 11(a) the termldquonearrdquo identifies the side that the crack started and ldquofarrdquoidentifies the side that the crack terminated After reachingthe fillet section of theweb debonding started at the near edgeof the CFRP patch While the crack moved towards the farside the debonding at the edge continued to grow Even afterthe crack reached to the far cut end the debonding remainedfairly stable Then the far edge of the CFRP patch startedto debond The reinforcement failed after around 50mm ofdebonding on both sides

Wu et al [38] studied the fatigue behaviour of sevensteel beams strengthened using four different strengthening


Crack along the weldCFRP plates removed

from the bond interface Adhesive in the interface between CFRP plate and

steel

(a) (b)

CFRP fabrics bondedwith Sikadur-330

Figure 12 Debonding interface of retrofitted beams using (a) CFRP plates (b) CFRP sheets [40]

materials During the test the fatigue crack always startedfrom the notch tip at midspan as the loading cycles increasedmainly because of the stress concentration in this areaInitially the cracks expanded slowly along the tension flangeand these cracks expanded at an increasing rate as theload cycled For the steel beam retrofitted by a weldedsteel plate the steel beam with plate fractured immediatelywhen the crack moved through the tension flange as shownin Figure 11(b)(i) and could not bear any further fatiguecycles For specimens reinforced with FRP composite platesextra loading cycles could still be sustained when the crackmoved through the tension flange As the load cycled thecrack expanded upwards on the web (Figure 11(b)(ii)) untildebonding failure (Figure 11(b)(iii)) No fatigue rupture wassustained by the reinforced composite plate Hence the useof a reinforced composite plate can significantly improve thefailure mode of the steel beam when compared to the useof a welded steel plate The difference in failure modes wasgenerally related to the fatigue strength and load-transferringmechanism of the reinforcement material However thefailure modes depend on the elastic modulus of the FRP andthe adhesive thickness [133]

Kim andHarries [37] investigated six beams intentionallydamaged by notching the tension flange of the beam toappraise the static and fatigue performance of the beamswith emphasis on local plasticity as well as the CFRP-steelinterface The retrofitted beam subjected to a high cyclicloading was observed to exhibit a large brittle fracture inthe web (Figure 11(c)) The form of brittle fracture beingaddressed had been termed ldquoconstraint-induced fracturerdquoand could occur without any noticeable fatigue crack growthand more importantly without any warning [134]

The FRP reinforcement ends and the regions wheregeometric discontinuities (cracks) take place are the mostsensitive zones to fatigue damage of the adhesive jointbecause of the stress concentration [28] Deng and Lee [55]found similar crack initiation and propagation in the CFRPreinforced steel beams except one test specimen which wasinvestigated with a 926 load range that caused it to debondsuddenly at one end after only 30 cycles Figure 11(d) Forthe other specimens the crack at each end grew quicklybut then almost stopped after one of the two cracks had

advanced past the midspan of the beams For all the platesthat had debonded the cracking initiated from the middle ofthe spew fillet and then propagated to the steel surface andglue interface at a 45∘ angle

Jiao et al [40] observed various failure modes for beamsstrengthened using CFRP plates and sheets A cut was madeand welded along the bottom tension flange of each beamFor unstrengthened beams a crack was initiated rapidly atthe tip of the cut once the crack propagated along the weldon the beam soffit throughout the whole tension flange Itshould be mentioned that the beam was considered to havefailed when it lost the ability to carry the maximum loadapplied in every cycle due to severe crack propagation Forstrengthened beams with CFRP plates debonding occurredin the steel surface and epoxy bond interface (Figure 12(a))Again failure happened between the plies of the sheetsin the strengthened beams as shown in Figure 12(b) Theobservation also confirmed that the performance of theCFRPplates to resist crack propagation was better than for theCFRP sheets under cyclic loading

The crack propagation rate depends on the stiffness ofthe FRP strip and largely on the prestressing force [73]Ghafoori et al [64] investigated the damaged steel beamsreinforced with the PUR and PBR methods and obtained asimilar load carrying capacity however the failure modeswere different In the PURmethod the CFRP plate slipped inthe mechanical anchorage system at the onset of the failurewhile the CFRP in the PBR method arrived at its tensilestrength in the cracked section and finally plate failure withdebonding (Figure 13)

The end anchoring technique mitigates the debonding ofthe FRP strips and maintains the prestressing force hencealso reducing the transfer length [68] Figure 14 illustrates theeffectiveness of the end anchoring technique in maintainingthe CFRP prestress The debonded CFRP strip rupture andthe strip end debonding model using FEA is shown inFigure 15

9 Conclusion

In this paper detailed reviews on the relevant researcheswere investigated systematically and carried out in respect of


CFRP slippage

(a)

CFRP failure and debonding

(b)

Figure 13 The failure modes for (a) PUR technique and (b) PBR technique

(a) (b)

Figure 14 Performance of end anchorage after failure for steel beam strengthened with prestressed strips to the cover plate (a) removableend anchor and (b) fixed end anchor [68]

(a) (b)

Figure 15 (a) Debonded strip rupture (b) end debonding model using FEA [68]

CFRPsteel strengthening techniques under fatigue Signifi-cant information and an explanation of the existing researchon the fatigue behaviour of FRP-strengthened steel structureshave been providedThe study also covered the surface treat-ment techniques adhesive curing and support conditionunder cyclic loading including fatigue performance crackpropagation and failure modes with FE simulation of thesteel bridge girders and structural elements

The following conclusions can be drawn from the presentstudy

(1) The remaining service life of bridge structures islimited by fatigue damage and in order to ensure thesafety of the steel bridges it is important to regularlycheck the structure to determine the existence offatigue cracks

(2) The application of HM HS UHM and prestressedCFRP and SW-BFRP strengthening composites notonly delays the initial crack reduces the crack growthrate and extends the fatigue life but also decreases thestiffness decay with residual deflection In additionstrengthened with prestressed CFRP had the beststrengthening effect

(3) The use of end anchorage prevented debonding of theCRRP strips at the beam ends by reducing the localinterfacial shear and peel stresses

(4) Epoxy adhesive curing is needed for potential FRPstrengthened structures Cyclic loading during adhe-sive curing can decrease the fatigue life of the rein-forced beam by reducing the bond strength


(5) Alteration in failure modes is mainly related to thefatigue strength and load-transferring mechanism ofthe strengthening material

(6) The prestressing force should be released very care-fully to avoid debonding of the strengthened steelbeams adhesively bonded with prestressed FRP

(7) Hybrid-joint nanoadhesive and carbon-flex can alsobe attractive for strengthening systems

The following recommendations are suggested for futureresearch in this area

(1) An effective surface treatment technique needs to bedeveloped for use in practice that can prevent failureat the adhesivesteel interface under cyclic loading

(2) A comparative analysis of the fatigue behaviour ofreinforced steel beams supported on a neoprene padand steel plate in a simply supported manner shouldbe investigated

(3) The effects of the thickness of the adhesive on thesystem require more investigation In addition theoptimum adhesive thickness should be studied toslow down delamination

(4) The length of optimum reinforcement should bedetermined to retard fatigue failure like end debond-ing

(5) The influence of different environmental conditionsmust be studied to find the real fatigue behaviour ofthe strengthening scheme

(6) The effect of an acidic or alkaline environment on thefatigue strength has not been studied yet

(7) The fatigue behaviour of damaged steel structuralelements reinforced with carbon-flex strengtheningtechniques should be investigated to enhance thecyclic load bearing capacity

(8) An appropriate rehabilitation method using FRPcomposites of welded steel structural elements undercyclic loading should be explored to retrofit weldedsteel bridge girders

(9) The application of nanoadhesives to FRPsteel bond-ed joints for rehabilitation of steel bridge girders toincrease the fatigue life has not been investigated yet

(10) The use of hybrid-joints to strengthen steel structuralelements with FRP requires further study

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

The authors gratefully acknowledge the support given byUniversity of Malaya (UM) for funding the work through theHigh Impact Research Grant H-16001-00-D000036

References

[1] J Bien L Elfgren and J Olofsson Sustainable Bridges Assess-ment for Future Traffic Demands and Longer Lives Wyda-wnictwo Edukacyjne Dolnosląskie Poland 2007

[2] E Ghafoori and M Asghari ldquoDynamic analysis of laminatedcomposite plates traversed by a moving mass based on a first-order theoryrdquoComposite Structures vol 92 no 8 pp 1865ndash18762010

[3] Y Kondo and K Okuya ldquoThe effect of seismic loading onthe fatigue strength of welded jointsrdquo Materials Science andEngineering A vol 468ndash470 pp 223ndash229 2007

[4] Y Lu L Mao and P Woodward ldquoFrequency characteristics ofrailway bridge response to moving trains with consideration oftrain massrdquo Engineering Structures vol 42 pp 9ndash22 2012

[5] S Hao ldquoI-35W bridge collapserdquo Journal of Bridge Engineeringvol 15 no 5 pp 608ndash614 2010

[6] K Nishikawa J Murakoshi and T Matsuki ldquoStudy on thefatigue of steel highway bridges in Japanrdquo Construction andBuilding Materials vol 12 no 2ndash4 pp 133ndash141 1998

[7] Y J Kim and P J Heffernan ldquoFatigue behavior of externallystrengthened concrete beams with fiber-reinforced polymersstate of the artrdquo Journal of Composites for Construction vol 12no 3 pp 246ndash256 2008

[8] L Van Den Einde L Zhao and F Seible ldquoUse of FRPcomposites in civil structural applicationsrdquo Construction andBuilding Materials vol 17 no 6-7 pp 389ndash403 2003

[9] N G Tsouvalis L S Mirisiotis and D N Dimou ldquoExperimen-tal and numerical study of the fatigue behaviour of compositepatch reinforced cracked steel platesrdquo International Journal ofFatigue vol 31 no 10 pp 1613ndash1627 2009

[10] H Liu R Al-Mahaidi and X-L Zhao ldquoExperimental study offatigue crack growth behaviour in adhesively reinforced steelstructuresrdquo Composite Structures vol 90 no 1 pp 12ndash20 2009

[11] Q Q Yu T Chen X L Gu X L Zhao and Z G Xiao ldquoFatiguebehaviour of CFRP strengthened steel plates with differentdegrees of damagerdquo Thin-Walled Structures vol 69 pp 10ndash172013

[12] A Shaat D Schnerch A Fam and S Rizkalla ldquoRetrofit of steelstructures using fiber-reinforced polymers (FRP) state-of-the-artrdquo in Proceedings of the Transportation Research Board (TRB)Annual Meeting CD-ROM (04-4063) 2004

[13] K A Harries A Varma S El-Tawil et al ldquoSteel-FRP compositestructural systemsrdquo in Proceedings of the 6th InternationalConference on Composite Construction in Steel and Concrete pp703ndash716 2011

[14] M Tavakkolizadeh and H Saadatmanesh ldquoFatigue strength ofsteel girders strengthened with carbon fiber reinforced polymerpatchrdquo Journal of Structural Engineering vol 129 no 2 pp 186ndash196 2003

[15] X-L Zhao and L Zhang ldquoState-of-the-art review on FRPstrengthened steel structuresrdquo Engineering Structures vol 29no 8 pp 1808ndash1823 2007

[16] H Suzuki ldquoFirst application of carbon Fiber reinforced polymerstrips to an existing steel bridge in Japanrdquo inAdvancedMaterialsfor Construction of Bridges Buildings and Other Structures2005

[17] R C Allan J Bird and J D Clarke ldquoUse of adhesives in repairof cracks in ship structuresrdquo Materials Science and Technologyvol 4 no 10 pp 853ndash859 1988


[18] I Grabovac R A Bartholomeusz and A A Baker ldquoCompositereinforcement of a ship superstructuremdashproject overviewrdquoComposites vol 24 no 6 pp 501ndash509 1993

[19] A Baker ldquoFibre composite repair of cracked metallic aircraftcomponentsrdquo in Proceedings of the Australian Aviation Sympo-sium lsquoInnovate or Enervatersquo Preprints of Papers p 107 1987

[20] A M P de Jesus J M T Pinto and J J L Morais ldquoAnalysis ofsolidwood beams strengthenedwithCFRP laminates of distinctlengthsrdquo Construction and Building Materials vol 35 pp 817ndash828 2012

[21] X Gu B Peng G Chen X Li and Y Ouyang ldquoRapidstrengthening of masonry structures cracked in earthquakesusing fiber composite materialsrdquo Journal of Composites forConstruction vol 16 no 5 pp 590ndash603 2012

[22] M Tavakkolizadeh and H Saadatmanesh ldquoRepair of damagedsteel-concrete composite girders using carbon fiber-reinforcedpolymer sheetsrdquo Journal of Composites for Construction vol 7no 4 pp 311ndash322 2003

[23] P T Curtis ldquoFatigue behaviour of fibrous composite materialsrdquoJournal of Strain Analysis for Engineering Design vol 24 no 4pp 235ndash244 1989

[24] Y Zheng L Ye X Lu and Q Yue ldquoExperimental study onfatigue behavior of tensile steel plates strengthened with CFRPplatesrdquo inProceedings of the 3rd International Conference on FRPComposites in Civil Engineering (CICE rsquo06) Miami Fla USADecember 2006

[25] Z S Wu J P Ou G Wu Y L Wang and W Q Feng ldquoAtype of continuous fiber-steel wires composite platerdquo ChinaNational Invention Patent Publication no CN101581133A 2009(Chinese)

[26] ZWu XWang and GWu ldquoAdvancement of basalt fiber com-posites towards infrastructural applications (Keynote paper)rdquoin Proceedings of the International Symposium on Innovation ampSustainability of Structures (ISISS rsquo11) October 2011

[27] S Moy ldquoEarly age curing under cyclic loading-an investigationinto stiffness development in carbon fibre reinforced steelbeamsrdquo in Proceedings of the 1st International Conference onAdvanced Polymer Composites for Structural Applications inConstruction Southampton UK April 2002

[28] D CNR 2022005 Guide for the Design and Constructionof Externally Bonded FRP Systems for Strengthening ExistingStructures National Research Council Rome Italy 2007

[29] B O Caglayan K Ozakgul and O Tezer ldquoFatigue life eval-uation of a through-girder steel railway bridgerdquo EngineeringFailure Analysis vol 16 no 3 pp 765ndash774 2009

[30] J Leander A Andersson and R Karoumi ldquoMonitoring andenhanced fatigue evaluation of a steel railway bridgerdquo Engineer-ing Structures vol 32 no 3 pp 854ndash863 2010

[31] O Buyukozturk O Gunes and E Karaca ldquoProgress on under-standing debonding problems in reinforced concrete and steelmembers strengthened using FRP compositesrdquo Constructionand Building Materials vol 18 no 1 pp 9ndash19 2004

[32] S C Jones and S A Civjan ldquoApplication of fiber reinforcedpolymer overlays to extend steel fatigue liferdquo Journal of Com-posites for Construction vol 7 no 4 pp 331ndash338 2003

[33] M Bocciarelli P Colombi G Fava and C Poggi ldquoFatigue per-formance of tensile steel members strengthened with CFRPplatesrdquo Composite Structures vol 87 no 4 pp 334ndash343 2009

[34] D Schnerch M Dawood S Rizkalla E Sumner and K Stan-ford ldquoBond behavior of CFRP strengthened steel structuresrdquoAdvances in Structural Engineering vol 9 no 6 pp 805ndash8172006

[35] B EN ldquoEurocode 3Design of Steel Structures-Part 1-9 FatiguerdquoEuropean Committee for Standardization (CEN) 2004

[36] B N B C BNBC Housing and Building Research InstituteBangladesh Standard and Testing Institute 1993

[37] Y J Kim and K A Harries ldquoFatigue behavior of damaged steelbeams repaired with CFRP stripsrdquo Engineering Structures vol33 no 5 pp 1491ndash1502 2011

[38] G Wu H Wang Z Wu H Liu and Y Ren ldquoExperimentalstudy on the fatigue behavior of steel beams strengthenedwith different fiber-reinforced composite platesrdquo Journal ofComposites for Construction vol 16 no 2 pp 127ndash137 2012

[39] U Loher B Mueller R Leutwiler and V Esslinger ldquoCFRP-strengthened aluminum structuresrdquo in Proceedings of the 17thInternational SAMPE Europe Conference on Success of Materialsby Combination pp 37ndash54 Basel Switzerland 1996

[40] H Jiao F Mashiri and X-L Zhao ldquoA comparative studyon fatigue behaviour of steel beams retrofitted with weldingpultruded CFRP plates and wet layup CFRP sheetsrdquo Thin-Walled Structures vol 59 pp 144ndash152 2012

[41] S Moy and F Nikoukar ldquoFlexural behaviour of steel beamsreinforced with carbon fibre reinforced polymer compositerdquo inProceedings of the Advanced Polymer Composites for StructuralApplications in Construction Proceedings of the 1st InternationalConference p 195 Southampton University Southampton UKApril 2002

[42] Y J Kim andG Brunell ldquoInteraction between CFRP-repair andinitial damage of wide-flange steel beams subjected to three-point bendingrdquo Composite Structures vol 93 no 8 pp 1986ndash1996 2011

[43] W Sebastian and C Zhang ldquoObservations from testing ofindeterminate FRP-plated steel beams on flexible supportsrdquoComposites Part B Engineering vol 45 no 1 pp 200ndash214 2013

[44] M A A Siddique and A A El Damatty ldquoImprovement oflocal buckling behaviour of steel beams through bonding GFRPplatesrdquo Composite Structures vol 96 pp 44ndash56 2013

[45] V Rance andU EvansCorrosion and Its Prevention at BimetallicContacts HM Stationary Office London UK 1958

[46] A Brown ldquoThe corrosion of CFRP-to-metal couples in salineenvironmentsrdquo in Proceedings of the 2nd International Confer-ence on Carbon Fibres pp 18ndash20 1974

[47] M Tavakkolizadeh and H Saadatmanesh ldquoGalvanic corrosionof carbon and steel in aggressive environmentsrdquo Journal ofComposites for Construction vol 5 no 3 pp 200ndash210 2001

[48] M Dawood S Rizkalla and E Sumner ldquoFatigue and over-loading behavior of steel-concrete composite flexural membersstrengthened with high modulus CFRP materialsrdquo Journal ofComposites for Construction vol 11 no 6 pp 659ndash669 2007

[49] M Dawood and S Rizkalla ldquoBond and splice behavior ofCFRP laminates for strengthening steel beamsrdquo in Proceedingsof the International Conference on Advanced Composites in Con-struction University of Bath Bath UK 2007

[50] A Yasin Seismic performance assessment of reinforced concreteframe structures [MS thesis] BUET Dhaka Bangladesh 2008

[51] L Hollaway and J Cadei ldquoProgress in the technique of upgrad-ing metallic structures with advanced polymer compositesrdquoProgress in Structural Engineering and Materials vol 4 pp 131ndash148 2002

[52] D Schnerch K Stanford E Sumner and S Rizkalla ldquoBondbehavior of CFRP strengthened steel bridges and structuresrdquo inProceedings of the International Symposium on Bond Behaviourof FRP in Structures (BBFS rsquo05) International Institute for FRPin Construction 2005


[53] A Monfared K Soudki and S Walbridge ldquoCFRP reinforcingto extend the fatigue lives of steel structuresrdquo in Proceedings ofthe Fourth International Conference on FRP Composites in CivilEngineering (CICE rsquo08) Zurich Switzerland 2008

[54] J Teng D Fernando T Yu and X Zhao ldquoTreatment of steelsurfaces for effective adhesive bondingrdquo in Advances in FRPComposites in Civil Engineering pp 865ndash868 2011

[55] J Deng and M M K Lee ldquoFatigue performance of metallicbeam strengthened with a bonded CFRP platerdquo CompositeStructures vol 78 no 2 pp 222ndash231 2007

[56] J Deng and M M K Lee ldquoBehaviour under static loadingof metallic beams reinforced with a bonded CFRP platerdquoComposite Structures vol 78 no 2 pp 232ndash242 2007

[57] D SchnerchMDawood S Rizkalla and E Sumner ldquoProposeddesign guidelines for strengthening of steel bridges with FRPmaterialsrdquo Construction and Building Materials vol 21 no 5pp 1001ndash1010 2007

[58] S Choudhury Effects of a debonding flew size and location on theflexural performance of FRP retrofitted steel sections [MS thesis]The University of Texas Arlington Tex USA 2007

[59] S A Hashim ldquoAdhesive bonding of thick steel adherends formarine structuresrdquo Marine Structures vol 12 no 6 pp 405ndash423 1999

[60] A F Harris and A Beevers ldquoThe effects of grit-blastingon surface properties for adhesionrdquo International Journal ofAdhesion and Adhesives vol 19 no 6 pp 445ndash452 1999

[61] A A El Damatty and M Abushagur ldquoTesting and modelingof shear and peel behavior for bonded steelFRP connectionsrdquoThin-Walled Structures vol 41 no 11 pp 987ndash1003 2003

[62] N Photiou L Hollaway and M Chryssanthopoulos ldquoAnultra-high modulus carbonglass fibre composite system forstructural upgrading of steelmembersrdquo inProceedings of the 2ndInternational Conference on FRPComposites in Civil Engineering(CICE rsquo04) pp 741ndash748 Adelaide Australia December 2004

[63] E Ghafoori A Schumacher and M Motavalli ldquoFatigue behav-ior of notched steel beams reinforced with bonded CFRP platesdetermination of prestressing level for crack arrestrdquo EngineeringStructures vol 45 pp 270ndash283 2012

[64] E Ghafoori M Motavalli J Botsis A Herwig and M GallildquoFatigue strengthening of damaged metallic beams using pre-stressed unbonded and bonded CFRP platesrdquo InternationalJournal of Fatigue vol 44 pp 303ndash315 2012

[65] Z Wu X Wang K Iwashita T Sasaki and Y HamaguchildquoTensile fatigue behaviour of FRP and hybrid FRP sheetsrdquoComposites B Engineering vol 41 no 5 pp 396ndash402 2010

[66] GWu ZWu Y Luo Z Sun andXHu ldquoMechanical propertiesof steel-frp composite bar under uniaxial and cyclic tensileloadsrdquo Journal of Materials in Civil Engineering vol 22 no 10Article ID 010010QMT pp 1056ndash1066 2010

[67] B M Phares T J Wipf F W Klaiber A Abu-Hawash andY-S Lee ldquoStrengthening of steel girder bridges using FRPrdquoin Proceedings of the Mid-Continent Transportation ResearchSymposium pp 1ndash12 2003

[68] F Vatandoost Fatigue behaviour of steel girders strengthenedwith prestressed CFRP strips [Master of Applied Science in CivilEngineering] The University of Waterloo Waterloo Canada2010

[69] P Colombi A Bassetti and A Nussbaumer ldquoDelaminationeffects on cracked steel members reinforced by prestressedcomposite patchrdquo Theoretical and Applied Fracture Mechanicsvol 39 no 1 pp 61ndash71 2003

[70] P Colombi A Bassetti and A Nussbaumer ldquoAnalysis ofcracked steel members reinforced by pre-stress compositepatchrdquo Fatigue and Fracture of Engineering Materials and Struc-tures vol 26 no 1 pp 59ndash66 2003

[71] K Soudki A Schumacher and M Motavalli ldquoFlexuralstrengthening of a steel beam with prestressed strips-prelimin-ary investigationrdquo in Proceedings of the 9th International Sym-posium on Fiber Reinforced Polymer Reinforcement for ConcreteStructures (FRPRCS rsquo09) Sydney Australia 2009

[72] D Fernando A Schumacher M Motavalli J Teng T Yu andE Ghafoori Fatigue Strengthening of Cracked Steel Beams withCFRP Plates American Society of Mechanical EngineeringVancouver Canada 2010

[73] B Taljsten C S Hansen and J W Schmidt ldquoStrengtheningof old metallic structures in fatigue with prestressed andnon-prestressed CFRP laminatesrdquo Construction and BuildingMaterials vol 23 no 4 pp 1665ndash1677 2009

[74] H Zhou and T L Attard ldquoRehabilitation and strength sus-tainability of fatigue damaged concrete-encased steel flexuralmembers using a newly developed polymeric carbon-fibercompositerdquo Composites Part B Engineering vol 45 no 1 pp1091ndash1103 2013

[75] H Zhou T L Attard B Zhao J Yu W Lu and L TongldquoExperimental study of retrofitted reinforced concrete shearwall and concrete-encased steel girders using a newCarbonFlexcomposite for damage stabilizationrdquo Engineering Failure Analy-sis vol 35 pp 219ndash233 2013

[76] H Zhou T L Attard Y Wang J Wang and F Ren ldquoReha-bilitation of notch damaged steel beams using a carbon fiberreinforced hybrid polymeric-matrix compositerdquo CompositeStructures vol 106 pp 690ndash702 2013

[77] M James ldquoOptimising the fatigue resistance of materialsrdquo 1992[78] C Society ldquoDesign guidance for strengthening concrete struc-

tures using fibre composite materialsrdquo Report of a ConcreteSociety Committee 2000

[79] F Nikouka M Lee and S Moy ldquoStrengthening of metallicstructures using carbon fibre compositesrdquo in Proceedings ofInternational Association for Bridge and Structural EngineeringSymposium (IABSE rsquo02) pp 121ndash127 Melbourne Australia2002

[80] P E Bourban S H McKnight S B Shulley V M Karbhariand J W Gillespie ldquoDurability of steelcomposites bonds forrehabilitation of structural componentsrdquo Proceedings of the 3rdMaterials Engineering Conference Infrastructure New Materialsand Methods of Repair pp 295ndash302 1994

[81] V M Karbhari and S B Shulley ldquoUse of composites for reha-bilitation of steel structuresmdashdetermination of bond durabilityrdquoJournal of Materials in Civil Engineering vol 7 no 4 pp 239ndash245 1995

[82] R A Barnes and G C Mays ldquoThe effect of traffic vibrationon adhesive curing during installation of bonded externalreinforcementrdquo Proceedings of the Institution of Civil EngineersStructures and Buildings vol 146 no 4 pp 403ndash410 2001

[83] S Moy ldquoCFRP reinforcement of steel beams adhesive cureunder cyclic loadrdquo in Proceedings of the 1st Asia-Pacific Confer-ence on FRP in Structures (APFIS rsquo07) pp 1019ndash1024 2007

[84] L Zhang L Hollaway J Teng and S Zhang ldquoStrengthening ofsteel bridges under low frequency vibrationsrdquo in Proceedings ofComposites in Civil Engineering (CICE rsquo06) Miami Fla USA2006


[85] T C Miller M J Chajes D R Mertz and J N HastingsldquoStrengthening of a steel bridge girder using CFRP platesrdquoJournal of Bridge Engineering vol 6 no 6 pp 514ndash522 2001

[86] K Narmashiri and M Z Jumaat ldquoReinforced steel I-beamsa comparison between 2D and 3D simulationrdquo SimulationModelling Practice and Theory vol 19 no 1 pp 564ndash585 2011

[87] E Ghafoori and M Motavalli ldquoAnalytical calculation of stressintensity factor of cracked steel I-beams with experimentalanalysis and 3D digital image correlation measurementsrdquo Engi-neering Fracture Mechanics vol 78 no 18 pp 3226ndash3242 2011

[88] C Czaderski and O Rabinovitch ldquoStructural behavior andinter-layer displacements in CFRP plated steel beamsmdashOpticalmeasurements analysis and comparative verificationrdquo Com-posites B Engineering vol 41 no 4 pp 276ndash286 2010

[89] C Czaderski K Soudki andMMotavalli ldquoFront and side viewimage correlation measurements on FRP to concrete pull-offbond testsrdquo Journal of Composites for Construction vol 14 no4 pp 451ndash463 2010

[90] Y Kawakam H Kanaji and K Oku ldquoStudy on application offield signature method (FSM) to fatigue crack monitoring onsteel bridgesrdquo Procedia Engineering vol 14 pp 1059ndash1064 2011

[91] Y J Xie H Xu and P N Li ldquoCrack mouth widening energy-release rate and its applicationrdquoTheoretical andApplied FractureMechanics vol 29 no 3 pp 195ndash203 1998

[92] W H Muller G Herrmann and H Gao ldquoElementary strengththeory of cracked beamsrdquo Theoretical and Applied FractureMechanics vol 18 no 2 pp 163ndash177 1993

[93] Y J Xie andXHWang ldquoApplication of Glowast-integral on crackedstructural beamsrdquo Journal of Constructional Steel Research vol60 no 9 pp 1271ndash1290 2004

[94] J G Teng T Yu and D Fernando ldquoStrengthening of steelstructures with fiber-reinforced polymer compositesrdquo Journalof Constructional Steel Research vol 78 pp 131ndash143 2012

[95] L Wenguang and C Guoping ldquoAnalysis on fatigue life ofcracked beam under forced vibrationrdquo in Proceedings of the2nd International Conference on Computer Engineering andTechnology (ICCET rsquo10) pp V546ndashV550 April 2010

[96] Z Xiulin and M A Hirt ldquoFatigue crack propagation in steelsrdquoEngineering FractureMechanics vol 18 no 5 pp 965ndash973 1983

[97] R Wang and A Nussbaumer ldquoModelling fatigue crack propa-gation of a cracked metallic member reinforced by compositepatchesrdquo Engineering Fracture Mechanics vol 76 no 9 pp1277ndash1287 2009

[98] P Paris and F Erdogan ldquoA critical analysis of crack propagationlawsrdquo Journal of Basic Engineering vol 85 p 528 1963

[99] S Walbridge ldquoFatigue analysis of post-weld fatigue improve-ment treatments using a strain-based fracture mechanicsmodelrdquo Engineering Fracture Mechanics vol 75 no 18 pp5057ndash5071 2008

[100] W Johnson and S Mall ldquoA fracture mechanics approach fordesigning adhesively bonded jointsrdquo ASTM STP vol 876 pp189ndash199 1985

[101] A Kinloch and S Osiyemi ldquoPredicting the fatigue life ofadhesively-bonded jointsrdquoThe Journal of Adhesion vol 43 no1-2 pp 79ndash90 1993

[102] S Erpolat I A Ashcroft A D Crocombe and M M Abdel-Wahab ldquoFatigue crack growth acceleration due to intermittentoverstressing in adhesively bonded CFRP jointsrdquo Composites AApplied Science andManufacturing vol 35 no 10 pp 1175ndash11832004

[103] I A Ashcroft ldquoA simple model to predict crack growth inbonded joints and laminates under variable-amplitude fatiguerdquoThe Journal of Strain Analysis for Engineering Design vol 39 no6 pp 707ndash716 2004

[104] J Lemaitre and R Desmorat Engineering Damage MechanicsDuctile Creep Fatigue and Brittle Failures Springer 2005

[105] P I Kattan and G Z Voyiadjis Damage Mechanics with FiniteElements Practical Applications with Computer Tools vol 1Springer Berlin Germany 2002

[106] P Raghavan and S Ghosh ldquoA continuum damage mechanicsmodel for unidirectional composites undergoing interfacialdebondingrdquoMechanics of Materials vol 37 no 9 pp 955ndash9792005

[107] B Bhattacharya and B Ellingwood ldquoContinuum damagemechanics analysis of fatigue crack initiationrdquo InternationalJournal of Fatigue vol 20 no 9 pp 631ndash639 1998

[108] M M A Wahab I A Ashcroft A D Crocombe and AD Crocombe ldquoPrediction of fatigue thresholds in adhesivelybonded joints using damage mechanics and fracture mechan-icsrdquo Journal of Adhesion Science and Technology vol 15 no 7pp 763ndash781 2001

[109] V Shenoy I A Ashcroft GW Critchlow and A D CrocombeldquoFracture mechanics and damage mechanics based fatiguelifetime prediction of adhesively bonded joints subjected tovariable amplitude fatiguerdquoEngineering FractureMechanics vol77 no 7 pp 1073ndash1090 2010

[110] D L Henry ATheory of Fatigue Damage Accumulation in SteelOhio State University 1953

[111] J A Bannantine J J Comer and J L Handrock Fundamentalsof Metal Fatigue Analysis vol 90 Prentice Hall EnglewoodCliffs NJ USA 1990

[112] B ANSI AISC 360-05-Specification for Structural Steel Build-ings AISC Chicago Ill USA 2005

[113] M Youssef ldquoAnalytical prediction of the linear and nonlinearbehaviour of steel beams rehabilitated using FRP sheetsrdquo Engi-neering structures vol 28 pp 903ndash911 2006

[114] H Zhou Y Wang Y Shi J Xiong and L Yang ldquoExtremely lowcycle fatigue prediction of steel beam-to-column connection byusing a micro-mechanics based fracture modelrdquo InternationalJournal of Fatigue vol 48 pp 90ndash100 2013

[115] A Pipinato C Pellegrino and C Modena ldquoFatigue assessmentof highway steel bridges in presence of seismic loadingrdquoEngineering Structures vol 33 no 1 pp 202ndash209 2011

[116] M S Cheung and W C Li ldquoProbabilistic fatigue and fractureanalyses of steel bridgesrdquo Structural Safety vol 25 no 3 pp245ndash262 2003

[117] P Colombi ldquoPlasticity induced fatigue crack growth retardationmodel for steel elements reinforced by composite patchrdquoTheo-retical and Applied Fracture Mechanics vol 43 no 1 pp 63ndash762005

[118] D L Chen BWeiss andR Stickler ldquoAmodel for crack closurerdquoEngineering FractureMechanics vol 53 no 4 pp 493ndash509 1996

[119] S Stoychev and D Kujawski ldquoMethods for crack opening loadand crack tip shielding determination a reviewrdquo Fatigue andFracture of Engineering Materials and Structures vol 26 no 11pp 1053ndash1067 2003

[120] J C Newman Jr ldquoA crack-closure model for predicting fatiguecrack growth under aircraft spectrum loadingrdquoASTM STP vol748 pp 53ndash84 1981

[121] G S Wang and A F Blom ldquoA strip model for fatigue crackgrowth predictions under general load conditionsrdquo EngineeringFracture Mechanics vol 40 no 3 pp 507ndash533 1991


[122] L C Hollaway and P R Head Advanced Polymer Compositesand Polymers in the Civil Infrastructure Elsevier Science 2001

[123] J Cadei T Stratford LHollaway andWDcukett StrengtheningMetallic Structures Using Externally Bonded Fibre-ReinforcedPolymers vol 595 CIRIA London UK 2004

[124] G Abed Effects of Temperature on the Adhesive Bonding inSteel Beams Reinforced with CFRP Composites University ofSouthampton 2012

[125] T Keller andM Schollmayer ldquoThrough-thickness performanceof adhesive joints between FRP bridge decks and steel girdersrdquoComposite Structures vol 87 no 3 pp 232ndash241 2009

[126] C-T Hoang-Ngoc and E Paroissien ldquoSimulation of single-lap bonded and hybrid (boltedbonded) joints with flexibleadhesiverdquo International Journal of Adhesion and Adhesives vol30 no 3 pp 117ndash129 2010

[127] M M Abdel Wahab ldquoFatigue in adhesively bonded joints areviewrdquo ISRNMaterials Science vol 2012 Article ID 746308 25pages 2012

[128] G Kelly ldquoQuasi-static strength and fatigue life of hybrid(bondedbolted) composite single-lap jointsrdquo Composite Struc-tures vol 72 no 1 pp 119ndash129 2006

[129] S Sam and M Shome ldquoStatic and fatigue performance ofweld bonded dual phase steel sheetsrdquo Science and Technologyof Welding and Joining vol 15 no 3 pp 242ndash247 2010

[130] F Moroni A Pirondi and F Kleiner ldquoExperimental analysisand comparison of the strength of simple and hybrid structuraljointsrdquo International Journal of Adhesion and Adhesives vol 30no 5 pp 367ndash379 2010

[131] S H Yoon andD G Lee ldquoIn situ crack propagationmonitoringin tubular adhesive joints containing quartz nano-particlesrdquoJournal of Adhesion Science and Technology vol 25 no 16 pp1973ndash1985 2011

[132] H Liu X Zhao and R Al-Mahaidi ldquoThe effect of fatigueloading on bond strength of CFRP bonded steel plate jointsrdquo inProceedings of the International Symposium on Bond Behaviourof FRP in Structures (BBFS rsquo05) Hong Kong China 2005

[133] A Pipinato C Pellegrino and C Modena ldquoFatigue behaviourof steel bridge joints strenghtened with FRP laminatesrdquoModernApplied Science vol 6 no 10 pp 1ndash15 2012

[134] K T Steudle and J Horsley AASHTO LRFD Bridge DesignSpecifications 6th edition 2012

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



additional weight because of increasing traffic volume andheavy traffic caused the fatigue failure The fatigue damagereported in Japan has been causedmainly by cyclic secondarystress or distortion-induced stress [6]

Fibre-reinforced polymer possesses outstanding advan-tages as a structural material including high strength anti-corrosion properties and high durability and is able torestore the lost capacity of damaged structures [7 8] FRPsheetsstrips are also effective in the strengthening of steelstructural elements to extend their fatigue lifetime and reducecrack propagation [9ndash11] if galvanic corrosion is preventedand sufficient bond is provided [12 13] Recent strengtheningprojects in the United States Switzerland the United King-dom and Japan showed that there was a great potential forusing carbonfibre-reinforced polymer (CFRP) to retrofit steelstructural elements [14ndash16] CFRP materials have been usedfor marine structures [17 18] and aerospace parts [19] andhave expanded to wood [20] masonry [21] steelconcretecomposite structures [22] and so on It is more frequentlyused for rehabilitation and strengthening of steel structuresthan other FRP materials due to its high strength In anycase CFRP is very tolerant to fatigue damage [23 24] Basalt-fibre-reinforced polymers (BFRP) have been increasinglyconsidered in civil infrastructures because of low cost andtheir excellent chemical and mechanical properties [25 26]

The Acton bridge in England [27] and the Ashlandbridge and 78385S092 bridge in the USA were retrofitted byapplying CFRP elements to the bottom of the girders thestress reduction on the original materials was observed toincrease the fatigue life [28] Several case studies of existingsteel railway bridges are presented in [29 30] Monitoringdata was used for dynamic response analysis and fatigue lifeevaluation It was observed that the stringers and the crossbeams run a high risk of fatigue damage The weakest pointin the reinforcing system of FRP elements to steel joints is thebond of the adhesive [31ndash33] The successful implementationof FRP composites of the strengthening systems is dependentupon the quality and integrity of the steel-composite jointand the effectiveness of the epoxy adhesive used [34] Thefatigue performance of the CFRP reinforcement details isnecessary to check if under realistic severe spectrum loadingthe adhesive joint performs better than the common fatiguesensitive details found on steel bridge girders in accordancewith Eurocode 3 categories 36lowast and 50lowast [35] or AmericanAssociation of State Highway and Transportation Officials(AASHTO) categories E1015840 and E [36]

The application of FRP composite for strengthening steelbridges and structural elements has become an efficientoption to meet the increased cyclic loads or repair due tocorrosion or fatigue cracking Therefore the objective of thisreview is to explore the potential FRP reinforcing techniquesto care for fatigue damaged steel structural elements Thispaper reviews the research related to CFRPsteel strength-ening techniques under fatigue Detailed knowledge andexplanation of the existing research concerning the fatiguebehaviour of HM HS UHM and prestressed CFRP and SW-BFRP strengthened steel structures are provided The studyalso covers the surface treatment techniques adhesive curingand support condition under cyclic loading including fatigue

performance crack propagation and failure modes with FEsimulation of the steel bridge girders and structural elementsFuture research gaps and recommendations are indicatedaccordingly

2 Surface Preparation and Treatment forSteel Beam Strengthening

21 Types of Notch To simulate the actual damage causedby corrosion and the expansion of fatigue cracks severalresearchers intentionally created notches of different geome-try in midspan or other positions on the tension flange of thebeams (Figure 1(a)) In addition the notch assists like a stressconcentrator [37] in the damage-sensitive regions [38 39] tocommence a vertical crack at the steel webThedifferent typesof notch can be categorized as follows

(a) rectangular notch on both edges (Figure 1(A))(b) U-shaped notch on both edges (Figure 1(B))(c) U-shaped notch through the whole tension flange

(Figure 1(C))(d) nonuniform notch (Figure 1(D))(e) uniform notch (Figure 1(E))

Usually the researchers cut a notch at midspan in thetension flange of the steel beams except Kim and Harries[37] To initiate the debonding of CFRP which was aimed atpropagating towards the right support they created a notchthrough the entire tension flange at a position 152mm to theright of midspan of the beams Jiao et al [40] welded the cutalong the tension flange soffit using the shielded metal arcwelding (SMAW) approach

The notch (C) through the whole tension flange issensitive to fatigue compared to the side notches (A) and(B) Accordingly notches (D) and (E) which go throughthe whole tension flange with part of the web in midspanare more sensitive to fatigue damage as the damage occurssuddenly

From the above categories of creating the notch it hasbeen revealed that when the notch spreads through theflange as well as the web of the steel beam the propagationof the cracks exists in the notch line of the flange andthe web Brittle fracture can happen in the case of fatigueThis is injurious for the structural element as no warningis given before failure In addition notches only createdin the flange can expect a retarding fracture with priorwarning However if the notch is given in the whole tensionflange there is also the possibility of brittle fracture underheavy repeated load Therefore for observing the properfatigue damage a rectangular and U-shaped notch on bothedges at midspan may be competently incorporated for thedevelopment of a standardized test Figure 1(b) illustratesthe stress characteristics of a strengthened steel beam undercyclic loading for incorporation of different notchesThe S-Ncurves show that the uniform notch comprising whole flangeand web has the least stress compared to others This clearlydemonstrates more fatigue life when the side rectangularnotch is incorporated in the flange


CFRPAdhesive

Flange

Web

CFRPFlange

Web09 mm

127 mm 127 mmNotch


Bottom flange

Notch

218 mm 218 mm

8 mm


Notch

152 mm

FlangeWeb

32 mm radius

t f=71

mm


FlangeWeb

15mm

5mm

1mm

5mm

Notch

tf


Flange

Web

Notch



(Wu et al 2012)


(a)

Figure 1 Continued


10

100

1000

10000 100000 1000000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles


(b)


Cover plate

Stiffener



























Notch at midspan

Adhesive


S127x45




119906= 297MPa and 03mm

thick





119906= 28GPa


Adhesive Notchx18



119906= 5413MPa


119906= 345GPa









Notch

Adhesive

S355J0 (ST 52-3)




119906=



Notch

Anchorage

S355J0 (ST 52-3)

(a) PUR system


Notch

AdhesiveAnchorage

S355J0 (ST 52-3)

(b) PBR system







4 Adhesive Curing









Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










CFRPAdhesive

Flange

Web

CFRPFlange

Web09 mm

127 mm 127 mmNotch


Bottom flange

Notch

218 mm 218 mm

8 mm


Notch

152 mm

FlangeWeb

32 mm radius

t f=71

mm


FlangeWeb

15mm

5mm

1mm

5mm

Notch

tf


Flange

Web

Notch



(Wu et al 2012)


(a)

Figure 1 Continued


10

100

1000

10000 100000 1000000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles


(b)


Cover plate

Stiffener



























Notch at midspan

Adhesive


S127x45




119906= 297MPa and 03mm

thick





119906= 28GPa


Adhesive Notchx18



119906= 5413MPa


119906= 345GPa









Notch

Adhesive

S355J0 (ST 52-3)




119906=



Notch

Anchorage

S355J0 (ST 52-3)

(a) PUR system


Notch

AdhesiveAnchorage

S355J0 (ST 52-3)

(b) PBR system







4 Adhesive Curing









Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










10

100

1000

10000 100000 1000000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles


(b)


Cover plate

Stiffener



























Notch at midspan

Adhesive


S127x45




119906= 297MPa and 03mm

thick





119906= 28GPa


Adhesive Notchx18



119906= 5413MPa


119906= 345GPa









Notch

Adhesive

S355J0 (ST 52-3)




119906=



Notch

Anchorage

S355J0 (ST 52-3)

(a) PUR system


Notch

AdhesiveAnchorage

S355J0 (ST 52-3)

(b) PBR system







4 Adhesive Curing









Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation





























Notch at midspan

Adhesive


S127x45




119906= 297MPa and 03mm

thick





119906= 28GPa


Adhesive Notchx18



119906= 5413MPa


119906= 345GPa









Notch

Adhesive

S355J0 (ST 52-3)




119906=



Notch

Anchorage

S355J0 (ST 52-3)

(a) PUR system


Notch

AdhesiveAnchorage

S355J0 (ST 52-3)

(b) PBR system







4 Adhesive Curing









Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












4 Adhesive Curing









Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Roller

Hydraulic jack

Steel plate

170


Adhesive

Support beam

Steel plate300 350180 5050


Strain gauge

Potentiometer

Unit mm

(a) [55]

Notch

Load cell

915 mm


763 mm152

Neoprene pad

flange width)

W150times 18

(100times 12mm full

(b) [37]

2800 mm

500

2000 mm

Ribbed stiffeners

LVDT Spreader beam

Strengtheningplates

Anchorages


188

70

70


(c) [38]

S127times 45

Strain gauge

Potentiometer

Unit mm

510 mm

CFRPAdhesive Notch

Loading block

200 mm 510 mm

Load cell



(d) [14]

Figure 3 Continued


395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










395 mm




Notch

Bottom loading rig

410 395

Stopper

Pin connected wheel


Strain gauge

Potentiometer

Unit mm

(e) [40]

CFRP

S355J0 (ST 52-3)

Roller

Notch

Hydraulic actuator

101050

Unit (mm)

Steel plateAdhesive

(f) [40 63]









Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Crack tip

Crack length


X

Y

Z

(a)

Anchorage


zone

CFRP plate

X

Y

Z

(b)









0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

50

100

150

200

0 5 10 15 20

Load

(kN

)





10

100

1000

1000 100000 10000000

Stre

ss ra

nge (

MPa

)

Number of cycles

ExperimentalFEA













10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











10

100

1000

1 10 100 1000 10000

Stre

ss ra

nge (

MPa

)










0

10

20

30

40

50

60

70

0 50000 100000 150000 200000 250000

Crac

k le

ngth

(mm

)

Number of cycles


+71

97

+0








0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70

Crac

k le

ngth

(mm

)


4 layers1 layer







0

15

30

45

60

75

90

105

0 8000 16000 24000 32000

Crac

k le

ngth

(mm

)

Number of cycles



1517192123252729313335

0 10000 20000 30000 40000 50000

Defl

ectio

n (m

m)

Number of cycles

Control15-F-S14-F-M

0-C-M35-F-M37-C-M





Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Far Bottom Near

(a)

(i) (ii) (iii)


Steel beam cracked

SW-BFRP debonded

(b)

(c) (d)










steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












steel

(a) (b)










9 Conclusion



CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










CFRP slippage

(a)


(b)


(a) (b)


(a) (b)

























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation


























Acknowledgment


References













































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation




































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









review article a review on strengthening steel beams using...

Documents