ruytenschildt bridge: field and laboratory testing
TRANSCRIPT
Engineering Structures 128 (2016) 111–123
Contents lists available at ScienceDirect
Engineering Structures
journal homepage: www.elsevier .com/ locate /engstruct
Ruytenschildt Bridge: Field and laboratory testing
http://dx.doi.org/10.1016/j.engstruct.2016.09.0290141-0296/� 2016 Elsevier Ltd. All rights reserved.
⇑ Corresponding author at: Delft University of Technology, Concrete Structures,Stevinweg 1, 2628 CN Delft, The Netherlands.
E-mail addresses: [email protected] (E.O.L. Lantsoght), [email protected] (Y. Yang), [email protected] (C. van der Veen), [email protected](A. de Boer), [email protected] (D.A. Hordijk).
Eva O.L. Lantsoght a,b,⇑, Yuguang Yang b, Cor van der Veen b, Ane de Boer c, Dick A. Hordijk b
aUniversidad San Francisco de Quito, Politecnico, Diego de Robles y Vía Interoceánica, Quito, EcuadorbDelft University of Technology, Concrete Structures, Stevinweg 1, 2628 CN Delft, The NetherlandscMinistry of Infrastructure and the Environment, Griffioenlaan 2, 3526 LA Utrecht, The Netherlands
a r t i c l e i n f o a b s t r a c t
Article history:Received 10 March 2016Revised 13 September 2016Accepted 15 September 2016Available online 23 September 2016
Keywords:AssessmentBeam testBending moment capacityBond propertiesField testPlain reinforcementShear capacitySlab bridge
A large number of existing reinforced concrete solid slab bridges in the Netherlands are found to be insuf-ficient for shear upon assessment. However, research has shown additional sources of capacity in slabbridges, increasing their total capacity. Previous testing was limited to half-scale slab specimens castin the laboratory. To study the full structural behavior of slab bridges, testing to failure of a bridge is nec-essary. In August 2014, a bridge was tested to failure in two spans. Afterwards, beams were sawn out ofthe bridge for experimental work in the laboratory and further study. Though calculations with currentdesign provisions showed that the bridge could fail in shear, the field test showed failure in flexure beforeshear. The experiments on the beams study the transition from flexural to shear failure and the influenceof the type of reinforcement on the capacity. The experimental results were compared to predictions ofthe capacity for the bridge slab and the sawn beams. These comparisons show that the current methodsfor rating of existing reinforced concrete slab bridges, leading to a sharper assessment, are conservative. Itwas also found that the application of plain bars instead of deformed bars does not increase the shearcapacity of beams.
� 2016 Elsevier Ltd. All rights reserved.
1. Introduction
1.1. Existing slab bridges in the Netherlands and code changes
The majority of the bridges in the Netherlands were built duringthe years following the Second World War. These bridges weredesigned for the live loads of that era, which are considerablylower than the current live loads (see NEN-EN 1991-2 + NA:2011[1]). In NEN-EN 1992-1-1:2005 [2] the shear capacity of a cross-section is also lower than in the previously used NEN 6720:1995[3].
The shear capacity for an element without axial load or pre-stressing and without shear reinforcement, according to NEN-EN1992-1-1:2005 [2] can be determined as follows:
VRd;c ¼ CRd;ckð100qlf ckÞ1=3bwdl P vminbwdl ð1Þ
k ¼ 1þffiffiffiffiffiffiffiffiffi200dl
s6 2:0 ð2Þ
with
VRd,c the design shear capacity in [kN];k the size effect factor, with dl in [mm];ql the flexural reinforcement ratio;fck the characteristic cylinder compressive strength of the con-crete in [MPa];bw the web width of the section in [m];dl the effective depth to the main flexural reinforcement in[mm].
According to the Eurocode procedures, the values ofCRd,c and vmin may be chosen nationally. The default values areCRd,c = 0.18/cc with cc = 1.5 in general and vmin (fck in [MPa]):
vmin ¼ 0:035k3=2f 1=2ck in ½MPa� ð3Þ
NEN-EN 1992-1-1:2005 Section 6.2.2 (6) accounts for the influenceof the shear span to depth ratio on direct load transfer. The contri-bution of a load applied within a distance 0.5dl 6 av 6 2dl from theedge of a support to the shear force VEd may be multiplied by thereduction factor b = av/2dl. In that clause of the code, the distanceav is considered as the distance between the face of the load andthe face of the support, or the center of the support for flexiblesupports.
Nomenclature
List of notationsav face-to-face distance between load and supportb member widthbedge distance between free edge and loadbpara effective width based on a parallel load spreading to the
straight casebskew effective width with horizontal load spreading under
45� from the far side of the wheel print to the face ofthe support
bstr effective width for a straight slabbw web widthdl effective depth to the longitudinal reinforcementfb the long-term concrete tensile strengthfck characteristic compressive strength (lower 5% bound)fcm average cube compressive strengthfcm,cyl average cylinder compressive strengthft average tensile strength of reinforcement steelfy average yield strength of reinforcement steelfyk characteristic yield strength of steelgk a parameter depending on the shear slendernessh member thicknessk size effect factorkh size effect factor from the Dutch NEN 6720:1995kk a factor to take continuity at the support into accountqw distributed load representing the self-weightvmin lower bound of the shear capacityvR,c average shear capacitywo reinforcement percentage, which needs to fulfill maxi-
mum and minimum valuesAc area of the concrete cross-sectionAo area used in NEN 6720:1995As area of steelCRd,c calibration factor in shear formula to determine the de-
sign shear capacityCRm,c calibration factor in shear formula to determine the
average shear capacityMcr calculated moment at cracking of the cross-section
Mdmax maximum absolute value of the design bending mo-ment in the member
Mtest maximum moment on the cross-section during theexperiment
Mu calculated moment at the ultimate of the cross-sectionMy calculated moment at yielding of the cross-sectionPcal,1 minimum of Py and Ps,ECPcal,2 minimum of Py and Ps,CSDPs,CSD calculated load for shear failure according to the Critical
Shear Displacement theoryPs,EC calculated load for shear failure according to NEN-EN
1992-1-1:2005 Section 6.2.2Ptot calculated maximum load at shear failure on tandemPtot,slab calculated maximum load at shear failure on tandem
after multiplying with slab increase factorPu maximum load during experimentPy calculated load at yielding in the critical cross-sectionVdmax maximum absolute value of the sectional shear in the
memberVEd design sectional shear force for the considered ULS load
combinationVRd,c design shear capacity according to the Eurocode provi-
sionsVu experimental sectional shearVu,CSD shear capacity according to Yang’s Critical Shear Dis-
placement theory (CSD)Vu,EC mean shear capacity according to the Eurocode provi-
sionscc partial factor for concrete materialu reinforcement bar diameterkv shear slendernessql reinforcement ratio of the longitudinal tensile reinforce-
mentsd design shear stress for the considered ULS load combi-
nations1 concrete contribution to the shear capacity
112 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
The shear capacity according to the previously used Dutch codeNEN 6720:1995 [3] can be verified with the following criterion:
sd 6 s1 ð4Þwith s1 the ultimate flexural shear capacity of the concrete elementwithout stirrups:
s1 ¼ 0:4f bkkkhffiffiffiffiffiffiwo
3p
P 0:4f b ð5Þwith
fb the concrete tensile strength, taken as the long-term tensilestrength [4]:
f bm ¼ 0:7ð1:05þ 0:05f cmÞ ð6Þkk for corbels and members at end supports where a compres-sion strut can be formed between the load and the support;kk = 1 for all cases, except:
kk ¼ 12gk
ffiffiffiffiffiffiffiffiffiffiffiffiffiAo
b� dl
3
sP 1 with
gk ¼ 1þ k2v if kv P 0:6gk ¼ 2:5� 3kv if kv < 0:6
and
kv ¼ Mdmax
dVdmaxð7Þ
kv the shear slenderness;Mdmax the maximum absolute value of the design bendingmoment in the member
Vdmax the maximum absolute value of the design sectional shearin the member;Ao is the smallest value of the area of the load or support, notexceeding b � dl;kh the size effect factor:
kh ¼ 1:6� h P 1:0 with h in ½m� ð8Þwo the reinforcement percentage, for members withoutprestressing:
wo ¼ 100As
b� dl6 2:0 and P 0:7� 0:5kv ð9Þ
sd the shear stress in the section, sd ¼ VEdb�dl
.
As a result of these code changes, upon assessment a large num-ber of existing Dutch reinforced concrete solid slab bridges arefound to be insufficient for shear [5].
1.2. Assessment by levels of approximation
In the fib Model Code 2010 [6], the concept Levels of Approxi-mation is introduced. Increasing the Level of Approximationincreases the computational time, but also results in a closer esti-mation of the capacity. Levels of Approximation are used for theModel Code shear and punching provisions [6].
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 113
Levels of Approximation are also used in The Netherlands forthe assessment of existing concrete structures, and in particularfor the shear assessment of slab bridges [7]. These levels are calledLevels of Assessment. The first Level of Assessment is the ‘‘QuickScan” [8], a conservative spreadsheet-based method that resultsin a ‘‘Unity Check”: the ratio of the sectional shear stress causedby the dead load, superimposed loads and live loads to the shearcapacity. If the Unity Check is larger than 1, the conclusion is notimmediately that the bridge does not have sufficient capacity,but that the analysis has to be repeated at Level of Assessment II.At this Level, the shear stress distribution over the width of thesupport is determined with a linear elastic finite element program.The peak shear stress is then averaged over 4dl (where dl is theeffective depth to the longitudinal reinforcement) [9] and com-pared to the shear capacity (same value as with Level of Assess-ment I) for the Level of Assessment II Unity Check. If the UnityCheck is again larger than 1, the procedure is repeated at Level ofAssessment III, which uses probabilistic analyses. Level of Assess-ment IV contains advanced non-linear finite element calculationsand proof loading. As more advanced Levels of Assessment requestmore time and labor, it is preferred that the lower Levels of Assess-ment, namely Levels I and II, are able to reach sufficient accuracy.
1.3. Past research on shear in slabs
Over the past few years, research has been carried out at DelftUniversity of Technology to study the behavior of reinforced con-crete slab bridges. Experiments were carried out on slab specimensunder concentrated loads close to supports [10–12]. It was con-cluded that slabs subjected to concentrated loads have additionalshear capacity as a result of transverse load redistribution [13]when compared to beams. These conclusions were also supportedby experimental evidence from the literature [14–20]. Theoreticalstudies led to the development of Yang’s Critical Shear Displace-ment theory [21,22]. For the shear analysis of slabs, the ExtendedStrip Model [23–25] was developed. Other suitable methods foradvanced shear analysis of slabs, at Levels of Approximation IIIand IV, include the use of probabilistic analyses, non-linear finiteelement analyses [26] or using the Critical Shear Crack Theory[27], taking into account the non-axis-symmetric nature of theproblem [28,29] when slab bridges are analyzed.
The additional capacity of slabs has been taken into account inLevel of Assessment I by the definition of an effective width inshear [7] at the slab support. In Level of Assessment II thisapproach resulted in the recommendation to distribute the peakstress over 4dl [9,30].
This paper looks at the highest Level of Assessment, which isfield testing, and aims at estimating the conservativeness of theLevel of Assessment I. In an exceptional case, an existing bridge,the Ruytenschildt Bridge in Friesland, was load tested to failure.Afterwards, beam specimens were sawn from the bridge andtested in the laboratory to further study the shear and flexuralcapacity of existing bridges. The field and laboratory testing ofthe same structure allow for a direct comparison between the dif-ferent types of tests, Usually, this comparison is not possible
Table 1Overview of past testing to failure on bridges.
Refs. Bridge name
Haritos et al. [50] Barr CreekAzizinamini et al. [51,52] Niobrara RiverMiller et al. [53,54] –Jorgenson and Larson [55] ND-18Bagge et al. [56,57] Kiruna
because laboratory tests tend to be simplifications and schematiza-tions of the reality. These tests are important, because field testingto failure of bridges is uncommon, and because the link betweenfield and laboratory tests is not typically made,
1.4. Past load testing to failure
In the past, only a limited number of bridges have been tested tofailure. An overview of those known by the authors is given inTable 1. It can be noted that the majority of these bridges were slabbridges, mostly resulting in a flexural failure. The testing of theThurloxton underpass is not included [31], because the test wascarried out after applying saw cuts over 1 m, so that only beambehavior and not slab behavior could be studied.
2. Description of Ruytenschildt Bridge
2.1. Introduction
The Ruytenschildt Bridge is located in the province of Friesland(the Netherlands), over a waterway connecting the Tjeuker Lake tothe Vierhuister Course and in the national road N924 connectingthe villages of Lemmer and Heerenveen. The bridge was built in1962, and was scheduled for demolition and replacement by abridge with a larger clearance for ship traffic, allowing for the pas-sage of taller boats. The carriageway was divided into two lanesand a bike lane.
2.2. Geometry
The structure was a solid slab bridge with five spans. At the midsupports, cross-beams cast integrally onto the piers were used, andat the end supports the deck is cast into the abutments; the bridgeis a fully integral bridge. The bridge had a skew angle of 18�. Thegeometry is shown in Fig. 1. The availability of at least one motorlane and one bicycle lane was needed during demolition andreconstruction. Therefore, the demolition of the bridge wasplanned in stages. In the first stage, 7.365 m of the bridge wasdemolished and the remaining 4.63 m served as one traffic lane.A saw cut between both parts was made, and the eastern part ofthe bridge of 7.365 m wide was tested to failure in spans 1 and2. A separate temporary bike bridge on pontoons was provided.The slab thickness was about 550 mm.
2.3. Material properties
62 cores were drilled to test the concrete compressive strength,showing that the average cube compressive strength wasfcm = 40 MPa [32]. Additional 31 concrete cores were drilled (invertical and horizontal direction) from the beam specimens inthe lab [33]. These additional tests provide a calibration factorfor the poor surface treatment of the previously tested cores,resulting in fcm = 63 MPa, which corresponds to a cylinder com-pressive strength fcm,cyl = 52 MPa. On the cores, the thickness ofthe asphalt layer was measured as 50 mm.
Type of bridge Failure mode
Slab bridge Flexural failureSlab bridge Flexural failureSlab bridge Punching failureSlab bridge Flexural failurePrestressed girder bridge Punching failure
7365 463512000
3000250
500250
600
35
6665
axis
exi
stin
gla
ne
Stage 1 Demolition East Side Existing
+2418
9000
9000
9000
9000
9000
W.S
.
+182
0+1
840
+184
0+1
820
+182
0+1
820
+182
0+1
820
+182
0+1
840
+184
0+1
820
+182
0+1
820
+184
0
+241
8
00090001
000990
0090
000009
0001
B
AA
egdirb gnitsixe sixA
B
Hee
renv
een
12000
700 7000 500 2975 8254200 3500
Echt
ener
brug
(a)
(b)
(c)
span 1
span 2
span 3
span 4
span 5
support 6
support 5
support 4
support 3
support 2
support 1
supp
ort 1
supp
ort 2
supp
ort 3
supp
ort 4
supp
ort 5
supp
ort 6
span
1sp
an 2
span
3sp
an 4
span
5
18o
N
unit: mm
Fig. 1. Overview of geometry of Ruytenschildt Bridge: (a) Tested part, cross-section; (b) side view; and (c) top view. Units: mm.
114 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 115
Reinforcement steel QR24 was used with diameters u = 22 mmand u = 19 mm for the longitudinal reinforcement and with diam-eter u = 12 mm for the transverse reinforcement. QR24 steel has acharacteristic yield strength fyk = 240 MPa. The reinforcement lay-out is shown in Fig. 2. Tensile tests on steel samples taken fromthe structure showed an average yield strength fy = 352 MPa andan average tensile strength ft = 435 MPa for the bars with a diame-ter u of 12 mm and fy = 309 MPa and ft = 360 MPa for u = 22 mm.The samples were taken from the bridge after testing, so that yield-ing of the steel could have occurred before determining the mate-rial properties. This suspicion seems to be confirmed by the limitedyielding plateau that was obtained in the measured load-deformation relationship of the steel samples. Past testing ofQR24 steel from a similar bridge gave fy = 282 MPa and ft = 402 -MPa [34].
2.4. State of Ruytenschildt Bridge prior to testing
In 2004 [35], a thorough inspection of the bridge was carriedout. The following structural problems were identified: crackingin the concrete slab over the entire depth of the cross-section(identified by the presence of water drops on the soffit of the slab),local signs of rebar corrosion, clogging of the drainage pipes, anddegradation of the timber sheet piles used at the abutment. Thedepth of carbonation was identified as 1–2 mm, which is wellwithin the concrete cover, and the chloride content was measuredon four samples, of which two had large amounts of chloride. Othersources of material degradation were not measured.
A few weeks prior to the testing of the bridge, a visual inspec-tion was carried out. This inspection identified again clogging ofthe drainage pipes, longitudinal and transverse cracks on the bot-tom and side faces of the slab, and a few locations with concretedamage caused by rebar corrosion.
Fig. 2. Reinforcement drawing of Ruytenschildt Bridge, showing spans 1, 2 and half of spUnits: mm.
3. Results of field testing of the Ruytenschildt Bridge
3.1. Test setup at bridge site
The geometry of the tandem load of NEN-EN 1991-2:2003 [36]was used for the test, with wheel prints of 400 mm � 400 mm, adistance along the width between the wheels of 2 m and an axledistance of 1.2 m. The face-to-face distance between axle andcross-beam was 2.5dl (1250 mm), which is the critical positionfor shear failure [5,37]. The distance between the saw cut lineand the first wheel was 800 mm in span 1 and 600 mm in span2. The tandem was placed in the obtuse angle, since this locationis critical for shear [38].
For a safe execution of the experiment, a steel load spreader, seeFig. 3, was applied over the span. Before the experiment, ballastblocks were placed on the load spreader, but the jacks were notyet extended, so that the slab was not loaded. During the experi-ment, the hydraulic jacks were gradually extended, and so the loadwas gradually transferred from the load spreader to the wheelprints positioned on top of the concrete slab. If large deformationscaused by failure would occur, the load on the jacks woulddecrease again thanks to this structure.
During the proof loading and testing to failure of theRuytenschildt Bridge, the structure was instrumented to studythe vertical and horizontal deformations, crack width, and toregister cracking activities with acoustic emission measurements.The vertical deformations were measured with linear variabledifferential transformers (LVDTs) and laser triangulation sensors.The deformations on the bottom surface, indicating the averagestrain over 1 m, were measured by LVDTs. The opening ofexisting cracks was followed with LVDTs. An interpretation ofthe measurements gathered during the field test and theirindication of the structure deviating from linear behavior is out-
an 3, and showing the location of the extracted beams. The structure is symmetric.
counter weight
load spreader
hydraulic jacks
supports
supports
Fig. 3. Overview of test setup on top of Ruytenschildt Bridge.
116 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
side of the scope of this paper, which studies the ultimate limitstate.
3.2. Measurements on Ruytenschildt Bridge
Two tests were carried out on the bridge: one test in span 1 anda second test in span 2, two days later. Several load cycles wereapplied during the field test, but only the last cycles of loading untilthe ultimate capacity are discussed here.
For span 2, the measurement scheme and the position of thetandem are given in Fig. 4. The reference for the LVDTs and laserswas drop wire. The loading scheme for the final step for span 1 isgiven in Fig. 5a, and for span 2 in Fig. 5b. The load-displacement
Fig. 4. Position of measurement equipment an
diagram of testing in span 2 as measured at one of the laser trian-gulation sensors is given in Fig. 6.
The maximum load during the test on span 1 was 3049 kN, butfailure was not achieved as the maximum load was determined bythe maximum available counter weight. Flexural cracking wasobserved in span 1, and no damage occurred at support 2. The testin span 1 did not cause additional precracking on span 2. For test-ing on span 2, two days later, additional counter weight wasordered. The maximum load was 3991 kN. Flexural failure wasachieved. A settlement at the pier of 15 mm right after achievingthe maximum load was observed. After removing all loading andmeasurement equipment, delayed recovery resulted in a finalresidual settlement of 8 mm.
d loading tandem for Span 2. Units: mm.
bstr
2.5dl
bpara
2.5dl
2.5dl
bedge
bskew
b edge
bedge
(a)
(b)
(c)
Fig. 7. Different possible effective widths for a skewed slab.
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0 2000 4000 6000 8000 10000
Load
(kN
)
time (s)
(b)
0
500
1000
1500
2000
2500
3000
3500
0 2000 4000 6000 8000
Load
(kN
)
time (s)
(a)
Fig. 5. Loading scheme to the ultimate: (a) span 1 and (b) span 2.
displacement at Laser 3 [mm]10 02 4 6 8 12 14 16
tota
l loa
d [k
N]
0
500
1000
1500
2000
2500
3000
3500
4000
4500
Fig. 6. Load-displacement diagram from span 2 as measured at laser 3.
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 117
3.3. Calculated and tested capacity
3.3.1. Shear capacity of Ruytenschildt BridgeTo predict the capacity of the tested cross-sections, calculations
were performed with average material parameters as given inSection 2.3. The characteristic shear capacity from NEN-EN1992-1-1:2005 Section 6.2.2. [2], see Eq. (1), can be converted intoan average shear capacity by using fcm and CRm,c = 0.15 [39]:
mR;c ¼ CRm;ckð100qlf cmÞ1=3 ð10ÞFor skewed slabs, the determination of the effective width in
shear is ambiguous and a direct application of the recommenda-tions for straight slabs is not possible. Three options have beenstudied: bstr, the effective width for a straight slab, bskew with hor-izontal load spreading under 45o from the far side of the wheelprint to the face of the support [7], and bpara based on a parallelload spreading to the straight case, as shown in Fig. 7.
The maximum total tandem load resulting in a Unity Check of 1was estimated: the maximum load is sought so that the resultingshear stress at the support from all occurring loads equals theshear capacity from Eq. (1). The recommendations from the slabshear experiments are taken into account [5]. However, knowledgeon how the plain bars affect the shear capacity is not available.Therefore, the same expressions as for deformed bars are used toestimate the shear capacity. Previous research on beams with plainbars [40–43] showed a higher shear capacity than for beams withdeformed bars. It can thus be thought that the provided values area lower bound of the real shear capacity of the bridge.
The maximum calculated tandem load is given as Ptot in Table 2.From the slab shear experiments [10], the 5% lower bound of the
ratio of the tested to predicted (Eq. (1)) shear capacity was foundto be 1.466 [44], mainly caused by transverse load redistribution.Multiplying Ptot by 1.466 gives Ptot,slab in Table 2. This multiplica-tion factor was derived in experiments on straight slabs. Skewedslabs, on the other hand, have larger stress concentrations in theobtuse corner, which could lead to smaller capacities [45]. There-fore, the full increase of the maximum load with the factor 1.466cannot immediately be extrapolated to skewed slabs, so that theshear capacity is estimated in between Ptot and Ptot,slab. The sixthrow of Table 2 shows the maximum load in the experiment.
For span 1, the maximum experimental load was smaller thanthe predicted maximum Ptot. If the slab effect is considered, thenthe maximum calculated load Ptot,slab is significantly higher thanthe experimental load. For span 2, similar conclusions can bedrawn. Since the slab failed in flexure, there is no indication ofits ultimate shear capacity.
Finally, it needs to be remarked that the influence of the inte-grally cast cross-beam was not accounted for in the calculations.The cross-beams induce some restraint that counteracts the shear.
Table 2Calculated shear and moment capacity of tested spans.
Span Span 1 Span 2
Shear capacity Ptot (kN) Ptot,slab (kN) Ptot (kN) Ptot,slab (kN)
bstr 3760 5512 4224 6192bpara 3236 4744 3608 5289bskew 4804 7043 5596 8204Experiment 3049 3991
Flexural capacity Span moment Support moment Span moment
Mcr (kN m) 1816 1690 1592My (kN m) 3489 5174 3412Mu (kN m) 4964 7064 4705Mtest (kN m) 4889 3306 4188
118 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
The longitudinal confinement caused by the support moments pos-sibly causes an increase in the shear capacity. A quantification ofthis effect is outside of the scope of this study.
3.3.2. Bending moment capacity of Ruytenschildt BridgeTo determine the moment at cracking Mcr, yielding My, and the
ultimate Mu, traditional beam analyses are carried out, see Table 2.The calculations are based on average material properties. Thecracking moment is based on the flexural tensile strength of theconcrete, calculated based on the ACI 318-14 [46] expression(function of the concrete compressive strength). For the momentat yielding, the stress-strain diagram of the concrete is approxi-mated with Thorenfeldt’s parabola. The ultimate moment Mu isconservatively based on ft = 360 MPa and is determined usingWhitney’s stress block for the concrete under compression. To findthe maximum experimental moment, Mtest, the two axles of theproof load tandem are applied as two point loads on a beam modelof five spans, and the self-weight is applied as a distributed load.
The Ruytenschildt Bridge is an integral bridge, and a momentwill develop at support 1. Because the rotational stiffness of sup-port 1 is difficult to estimate, it was conservatively estimated asa hinge. In reality, a support moment develops which decreasesthe span moment, explaining why failure did not occur at Mtest
for testing in span 1.For span 2, Mtest lies in between My and Mu, which corresponds
with the experiment: span 2 failed in flexure in the span; yieldingof the steel was observed but crushing of the concrete was notachieved.
The Ruytenschildt Bridge had identical span lengths for all fivespans. Most continuous bridges have shorter end spans, whichwould result in different values for shear and moment.
4. Description of beams sawn from the Ruytenschildt Bridge
4.1. Introduction
Before the remaining part of the bridge was demolished, beamswere extracted from the spans that were not tested in the field sothey could be tested in the laboratory. On these beams, the trans-verse redistribution that occurs in slabs can be excluded. Theeffects of the skew angle are excluded as well. Thus, the measuredshear or flexural behavior can be directly compared with the codeprovisions for beams, which are the basis of the calculations pre-sented previously.
Two topics that were of interest for further studies and thatwere analyzed in the beam tests are the effect of plain reinforce-ment bars on the shear capacity, the occurring failure mode (isshear failure even realistic?), and the additional deformationcapacity after the yielding moment of a critical section is reached.Studying these aspects may lead to an improved understandingand assessment of existing bridges.
4.2. Geometry of the sawn beams
Three beams, RSB01-RSB03, of 6 m were sawn from span 1. Thepositions of the beams in the deck are indicated in Fig. 2. Theintended width of the specimens was 500 mm for RSB01 andRSB02, and 1000 mm for RSB03. The sawing operation at the siteof the viaduct was not a precise process. As a result, the width ofthe beamswas larger than originally intended. Therefore, the actualcross-sections of the specimens were measured at five positions: at0.6 m, 1.8 m, 3.0 m, 4.2 m and 5.4 m from the end of the specimens,resulting in an overview of the cross-sections as given in Fig. 8a.
The asphalt layer was kept on the specimens. The average thick-ness of the layer is 50 mm. This layer was only removed at theloading plate (except for RSB03A) and the top surface was leveledwith high strength mortar. This treatment ensures that the poormechanical properties of the asphalt will not influence the loadingprocess. On the remaining parts of the beam, the asphalt layer waskept to maintain the flexural stiffness of the original bridge and theoriginal dead load state that included the superimposed dead load.In order to check the influence of removing the asphalt layer, intest RSB03A the asphalt layer is kept.
According to Fig. 2, the longitudinal reinforcement configurationis repeated at every 270 mm. It is grouped into 4 layers; the detailsof each layer are shown in Fig. 8b. The bars are arranged in thedirection of the shorter edge of the deck. The saw cut lines of thebeams were in the transverse direction, see Fig. 2. Thus, the longi-tudinal bars were not aligned with the saw cut. The positions ofthe bars with respect to the saw cut in the longitudinal directionare not consistent among the beams, therefore these positions haveto be checked individually. The actual positions of the saw cuts incomparison with the longitudinal reinforcement layout are indi-cated in Figs. 8b and 9, determining the positions of the support.Take RSB01 for example: the supports were placed closer to endB, because the Ø19 rebar in layer 4 is closer to the B end.
Similarly, the availability of the reinforcement is checked in thewidth direction. For all three specimens, the sawing operationaffects the longitudinal bars at the edges, see Fig. 8a and b. Thesepartial bars are not taken into account for the shear capacity. Forthe flexural capacity of sections more than 1 m away from thedamaged rebar, these bars are accounted for. The calculatedanchorage length according to NEN-EN 1992-1-1:2005 Eq. 8.3 [2]is 452 mm. The applied anchorage length (1 m) is more than dou-ble the code requirement to compensate for the potentially weakerbond between plain rebar and concrete. Similar considerationsdetermine the positions of the supports. Based on the reinforce-ment configuration sketched in Fig. 8b, the sectional and reinforce-ment properties are given in Fig. 9 and Table 3.
4.3. Test setup
The specimens are simply supported with a span of 5 m andloaded by a point load. The position of the point load varies
(a)
(b)
Fig. 8. Properties of sawn beams: (a) Measured cross-sections of specimens. Units: mm. (b) Rebar configuration in the beam cross-sections. Bar diameters in mm.
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 119
according to the type of test. For the bending tests (RSB01F andRSB03F, with ‘‘F” for flexural test), the point load is located atmidspan. For the shear test, the loading position is at 1.25 mfrom the support in RSB02A and RSB02B and at 1.3 m in RSB03A,with ‘‘A” or ‘‘B” denoting the support close to which was tested.The loading geometry of all tests is indicated in Fig. 9. Duringthe experiments, the magnitude of the load, vertical deflectionsand crack widths were measured. In all tests, the deflection atthe loading point was measured from the bottom of thespecimen (on both sides of the beam). The support deformationwas compensated from the measurements. A typical load-deflection relationship of specimens failing in flexure and inshear are given in Fig. 10a and b respectively. Acoustic emissionmeasurements were used to study crack development andpropagation.
In total 5 tests were executed on 3 specimens, among which twotests were carried out on specimens RSB02 and RSB03. On RSB02,the first test was RSB02A. By the end of that test, a flexural shearcrack was formed, nevertheless, the specimen did not collapse afterthe formation of the crack. Instead, a yielding plateau was observedin the load-deflection relationship. In order to guarantee the struc-tural integrity for the test RSB02B, the test was stopped with lim-ited deflection. The loading point was moved to the B end, wherea flexural shear failure was found. In the case of RSB03, similarapproach was taken. The first test of the specimen was a flexuraltest RSB03F. Afterwards, the shear test was executed. During thewhole experimental program, the position of the supports wasnot changed. According to [21], it is assumed that this will not affectthe crack pattern of the specimen, and eventually the shear capac-ity. This is further validated in the experiments reported in [39,47].
Fig. 9. Test setups of specimen RSB01-RSB03. Units: mm.
Table 3Properties of critical cross-sections of the beams.
RSB01F RSB02A RSB02B RSB03F RSB03A
dl (mm) 503.0 515.5 520.0 521 515Ac (m2) 0.290 0.297 0.3066 0.596 0.537Rebar 4Ø22 + 4Ø19 4Ø22 + 4Ø19 4Ø22 + 5Ø19 9Ø22 + 8Ø19 7Ø22 + 8Ø19ql 0.91% 0.89% 0.96% 0.95% 0.92%
120 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
5. Test results of beams
5.1. Overview of results
The theoretical yield moment My, the shear capacity accordingto the Eurocode provisions (Eq. (1)) Vu,EC, and according to Yang’sCritical Shear Displacement theory (CSD), Vu,CSD [21,22] are listed
in Table 4. The calculated values are determined by using theaverage cross-sectional properties from Fig. 8a. The experimentalsectional shear Vu and maximum load Pu are also given in Table 4.The critical shear crack is assumed at 1 m from the closest support.Py is the calculated load at yielding in the critical cross-section.Ps,EC is the calculated load for shear failure according to NEN-EN1992-1-1:2005 Section 6.2.2., see Eq. (1). Ps,CSD is the calculated
0 5 10 150
100
200
300
400
500
600
700
load
[kN
]
(a)
0 5 100
100
200
300
400
500
600
700
load
[kN
]
(b)A30BSRF30BSR
Fig. 10. Load-deflection relationship of RSB03F and RSB03A.
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 121
load for shear failure according to the Critical Shear Displacementtheory proposed in [21,22]. Pcal,1 is the minimum of Py and Ps,EC andPcal,2 of Py and Ps,CSD.
5.2. Shear behavior
In both RSB02A and RSB02B, an inclined crack developed in theshear span. The formation of this crack did not result in a drop ofthe capacity of the specimen. In RSB02B, an inclined crack wasobserved before yielding of the longitudinal reinforcement. Thetest was stopped by then to ensure the structural integrity forthe test RSB02A. After that, a second test at end B was done asthe continuation of RSB02B. The same loading position did notresult in a significant additional deflection. Failure then occurredby crushing of the compression strut. In RSB03A, the shear spanwas increased to 1.3 m, resulting in failure by forming an inclinedcrack in the shear span.
In the literature [42,48], larger shear capacities are found forbeams reinforced with plain bars than for beams with deformedbars. However, when studying the shear tests presented here, theexperimental shear capacities are not significantly different fromthe calculated shear capacities based on models for specimenswith deformed bars. Similar conclusions were drawn in the pastfrom tests on slabs reinforced with plain bars as compared todeformed bars, subjected to concentrated loads close to supports[49]. The expression given by Eurocode Eq (1), turns out to give areasonable estimation of the test results, with an average ratio oftested to predicted values of 1.2 and a coefficient of variation of5.2%. This difference partially contributes to the underestimationof the ultimate capacity of the bridge during the test preparationphase.
A comparison with Yang’s Critical Shear Displacement Theoryshows an average ratio of tested to predicted results of 0.95 anda coefficient of variation of 4.8%.
Table 4Comparison of test results and model predictions.
My (kN m) Vu,EC (kN) Vu,CSD (kN) Vu (kN) Py (
RSB01F 369.2 245.4 275.9 149.9 275RSB02A 378.3 254.4 327.5 289.1 376RSB02B 422.5 260.3 338.8 330.7a 423RSB03F 819.0 480.3 468.2 326.5 617RSB03A 809.3 469.9 603.1 546.2 792
a The load dropped due to the formation of an inclined crack. The test was stopped to kexecuted at the same load position. It turned out that the load level could be increaseddetermined for Pu = 424.2 kN.
b Flexural shear failure.c Flexural failure.
When linking the results of the beam tests to the capacity of thebridge tested in the field, it must be noted that the longitudinalconfinement, effects of continuity, and the transverse redistribu-tion capacity of the slab will increase the shear capacity beyondthe capacity as quantified on the beam. The results of the beam testand the field test can be linked, as similar shear spans were used inthese tests. These effects also result in a shift of the failure modefor existing slab bridges: flexural failures become more probableand shear failures less probable.
5.3. Flexural behavior
The comparison between experimental and calculated valuesshows that the prediction of the yielding moment is accurate, witha difference of maximum 2%. Additionally, it was confirmed thatbeams with plain bars have large rotational capacities. In RSB01F,the residual deformation was more than 125 mm (250 mm/2, seeFig. 11c and d), which is 1/40 of the span length. The test had tobe stopped because the maximum displacement of the actuatorwas reached. The load level before the stop of the test was still271 kN, while the peak load was 274 kN.
6. Recommendations
From the field testing, the following recommendations can begiven:
� Further research is necessary to identify the effect of the skewangle on the shear capacity. Meanwhile, the uncertainties canbe covered by applying the different effective widths fromFig. 7, and to define a range of values between which shear fail-ure could occur.
� An initial shear assessment can be carried out with the QuickScan method [8].
kN) Ps,EC (kN) Ps,CSD (kN) Pu (kN) Pu/Pcal,1 Pu/Pcal,2
.8 466.8 443.4 275.8c 1.00 1.00
.2 322.5 420.0 368.7c 1.14 0.98
.4 331.2 435.4 415.8a 1.26 0.98
.3 914.1 889.9 606.6c 0.98 0.98
.0 603.7 783.7 706.7b 1.17 0.90
eep the integrity of the specimen for another test at end A. Later, a second test wasfurther to 424.2 kN when the compression zone of the specimen failed. Thus, Vu is
(a)
25 cm
(b)
(e)(d)(c)
Fig. 11. Photographs of beams tests: (a and b) rebar damage of RSB03, end A; (c) crack caused by flexural failure of RSB01; and (d and e) residual deformation of RSB01 afterfailure.
122 E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123
From the laboratory testing, the following recommendations canbe given:
� The shear capacity can be predicted with the shear provisionsfrom Eurocode 2 [2], as well as with Yang’s Critical ShearDisplacement Theory [21,22].
� In structures with plain reinforcement bars with a low yieldstrength, shear failures can occur, which was contrary to theexpectation. Therefore, further experimental work on theflexural and shear capacity of beams with low levels of rein-forcement is under development.
� The shear capacity of elements with plain reinforcement barscan be determined in the same way as the shear capacity of ele-ments with ribbed bars.
7. Summary and conclusions
This paper presented a test to failure executed in the field on areinforced concrete slab bridge from 1962, as well as on threebeams sawn from the bridge. A literature review showed that col-lapse tests on reinforced concrete slab bridges are scarce. Thisobservation motivated the authors to carry out a well-instrumented collapse test on an existing bridge, and extend thisresearch with the testing of beams in the laboratory.
Two spans were tested in the field. In the first span, flexural dis-tress was observed, but failure did not occur. One of the reasonswhy the flexural capacity was higher in span 1 than predicted, isthat the tested bridge was an integral bridge. The effect of the sup-port moment resulted in a higher moment capacity in the spanthan when assuming hinges. In the second span, more counter-weight was provided and a flexural failure could be achieved.Despite the effect of the different boundary conditions, the flexuralcapacity of the tested bridge could be estimated satisfactorily.However, the test showed that the shear capacity of the bridgedeck was higher.
The beams tested in the lab failed in shear and flexure. As such,shear failures cannot be excluded for elements with plain rein-forcement. Moreover, a larger shear capacity for beams reinforced
with plain bars than with deformed bars was not observed in thelab experiments.
A final conclusion of the presented research is that the currentrating procedures at Level of Assessment I (the Quick Scan method)are conservative for existing reinforced concrete slab bridges.
Acknowledgements
The authors wish to express their gratitude and sincere appre-ciation to the Province of Friesland and the Dutch Ministry ofInfrastructure and the Environment (Rijkswaterstaat) for financingthis research work. The contributions and help of our colleagues A.Bosman, S. Fennis and P. van Hemert, of the contractor de Boer ende Groot, and of Mammoet, responsible for applying the load, arealso gratefully acknowledged.
References
[1] Code Committee 351001. Eurocode 1 – actions on structures – part 2: trafficloads on bridges, NEN-EN 1991–2/NA:2011. Delft (The Netherlands): CivilEngineering Center for Research and Regulation, Dutch NormalizationInstitute; 2011. p. 38.
[2] CEN. Eurocode 2: design of concrete structures – part 1–1 general rules andrules for buildings. NEN-EN 1992-1-1:2005. Brussels (Belgium): ComitéEuropéen de Normalisation; 2005. p. 229.
[3] Code Committee 351001. NEN 6720 technical guidelines for structures,provisions concrete TGB 1990 – Structural requirements and calculationsmethods (VBC 1995). Delft (The Netherlands): Civieltechnisch centrumuitvoering research en regelgeving, Nederlands Normalisatie-instituut; 1995[in Dutch].
[4] CUR. CUR report 94–13, 1994, background to VBC 1990. Gouda (TheNetherlands): Civieltechnisch Centrum Uitvoering Research en Regelgeving;1994. p. 162 [in Dutch].
[5] Lantsoght EOL, van der Veen C, de Boer A, Walraven JC. Recommendations forthe shear assessment of reinforced concrete slab bridges from experiments.Struct Eng Int 2013;23:418–26.
[6] FIB. Model code 2010: final draft. Lausanne: International Federation forStructural Concrete; 2012.
[7] Lantsoght EOL, de Boer A, van der Veen C, Walraven J. Effective shear width ofconcrete slab bridges. Proc Inst Civil Eng – Bridge Eng 2015;168:287–98.
[8] Vergoossen R, Naaktgeboren M, Hart MT, De Boer A, Van Vugt E. Quick scan onshear in existing slab type viaducts. In: International IABSE conference onassessment, upgrading and refurbishment of infrastructures. Rotterdam (TheNetherlands); 2013. p. 8.
E.O.L. Lantsoght et al. / Engineering Structures 128 (2016) 111–123 123
[9] Lantsoght EOL, De Boer A, van der Veen C. Determination of distribution widthfor shear stresses at support in reinforced concrete slabbridges. Austria: EuroC. St. Anton am Arlberg; 2014. p. 9.
[10] Lantsoght EOL, van der Veen C, Walraven JC. Shear in one-way slabs under aconcentrated load close to the support. ACI Struct J 2013;110:275–84.
[11] Lantsoght EOL, van der Veen C, De Boer A, Walraven J. Influence of width onshear capacity of reinforced concrete members. ACI Struct J2014;111:1441–50.
[12] Lantsoght EOL, van der Veen C, Walraven J, de Boer A. Experimentalinvestigation on shear capacity of reinforced concrete slabs with plain barsand slabs on elastomeric bearings. Eng Struct 2015;103:1–14.
[13] Lantsoght EOL, van der Veen C, de Boer A, Walraven J. Transverse loadredistribution and effective shear width in reinforced concrete slabs. Heron2014 [in press, 29 pp.].
[14] Graf O. Experiments on the capacity of reinforced concrete slabs under aconcentrated load close to a support. Deutscher Ausschuss für Eisenbeton1933;73:10–6 [in German].
[15] Regan PE. Shear resistance of concrete slabs at concentrated loads close tosupports. London (United Kingdom): Polytechnic of Central London; 1982. p.24.
[16] Furuuchi H, Takahashi Y, Ueda T, Kakuta Y. Effective width for shear failure ofRC deep slabs. Trans Jpn Conc Inst 1998;20:209–16.
[17] Reißen K, Hegger J. Experimental investigations on the effective width forshear of single span bridge deck slabs. Beton- und Stahlbetonbau2013;108:96–103 [in German].
[18] Reißen K, Hegger J. Experimental Investigations on the shear capacity of RCcantilever bridge deck slabs under concentrated loads – influences of moment-shear ratio and an inclined compression zone. In: European bridge conference.Edinburgh (UK); 2015. p. 13.
[19] Rombach GA, Latte S. Shear resistance of bridge decks without shearreinforcement. In: International FIB symposium 2008. Amsterdam (TheNetherlands): Fédération Internationale du Béton; 2008. p. 519–25.
[20] Natário F, Ruiz MF, Muttoni A. Shear strength of RC slabs under concentratedloads near clamped linear supports. Eng Struct 2014;76:10–23.
[21] Yang Y. Shear behaviour of reinforced concrete members without shearreinforcement - a new look at an old problem [Ph.D. Thesis]. Delft: DelftUniversity of Technology; 2014.
[22] Yang Y, Den Uijl JA, Walraven J. The critical shear displacement theory: on theway to extending the scope of shear design and assessment for memberswithout shear reinforcement. Struct Conc 2016 [available online ahead ofprint: 25 pp.].
[23] Lantsoght EOL, van der Veen C, de Boer A. Predicting the shear capacity ofreinforced concrete slabs subjected to concentrated loads close to supportswith the modified bond model. IABSE 2014, engineering for progress. Madrid,Spain: Nature and People; 2014. 2014.
[24] Lantsoght EOL, van der Veen C, de Boer A. Modified bond model for shear inslabs under concentrated loads. In: Stang H, Braestrup M, editors. FIBsymposium. Copenhagen: Concrete Innovation and Design; 2015. p. 10.
[25] Lantsoght EOL, Van der Veen C, De Boer A, Alexander SDB. Bridging the gapbetween one-way and two-way shear in slabs. In: ACI special publication forthe international punching symposium; 2016. p. 20 [in review].
[26] Belletti B, Damoni C, Hendriks MAN, de Boer A. Analytical and numericalevaluation of the design shear resistance of reinforced concrete slabs. StructConc 2014;15:317–30.
[27] Muttoni A, Fernández Ruiz M. Shear strength in one- and two-way slabsaccording to the critical shear crack theory. In: International FIB symposium2008. Amsterdam (The Netherlands): Fédération Internationale du Béton;2008. p. 559–63.
[28] Sagaseta J, Muttoni A, Ruiz MF, Tassinari L. Non-axis-symmetrical punchingshear around internal columns of RC slabs without transverse reinforcement.Magaz Conc Res 2011;63:441–57.
[29] Falbr J. Shear redistribution in solid concrete slabs [M.Sc. Thesis]. Delft: DelftUniversity of Technology; 2011.
[30] Lantsoght EOL, de Boer A, Van der Veen C, Walraven JC. Peak shear stressdistribution in finite element models of concrete slabs. In: Zingoni A, editor.Research and applications in structural engineering, mechanics andcomputation. Cape Town (South Africa): Taylor and Francis; 2013. p. 475–80.
[31] Cullington DW, Daly AF, Hill ME. Assessment of reinforced concrete bridges:collapse tests on Thurloxton underpass. Bridge Manage 1996;3:667–74.
[32] Lantsoght EOL. Ruytenschildt Bridge: analysis of test results at the ultimatelimit state [Stevin Report 25.5-15-05]. Delft (The Netherlands): DelftUniversity of Technology; 2015. p. 89 [in Dutch].
[33] Yang Y. Experimental studies on the structural behaviours of beams fromRuytenschildt Bridge. Stevin Report 25.5-15-09, 2015. p. 61.
[34] Yang Y, Den Uijl JA, Dieteren G, de Boer A. Shear capacity of 50 years oldreinforced concrete bridge deck without shear reinforcement. In: 3rd FIBinternational congress. Washington (DC, USA): FIB; 2010. p. 12.
[35] de Pee J. Inspection report Ruytenschildt Bridge. BV: Nebest; 2004 [in Dutch].[36] CEN. Eurocode 1: actions on structures – part 2: traffic loads on bridges, NEN-
EN 1991-2:2003. Brussels (Belgium): Comité Européen de Normalisation;2003. p. 168.
[37] Rijkswaterstaat. Guidelines assessment bridges – rating of structural safety ofan existing bridge at repair, usage and disapproval; 2013. p. 117 [in Dutch].
[38] Cope RJ. Flexural shear failure of reinforced-concrete slab bridges. Proc InstCivil Eng Part 2-Res Theory 1985;79:559–83.
[39] Yang Y, den Uijl JA. Report of experimental research on shear capacity ofbeams close to intermediate supports [Stevin Report 25.5-11-10]. Delft: DelftUniversity of Technology; 2011.
[40] Kani GNJ. The riddle of shear failure and its solution. ACI J Proc1964;61:441–67.
[41] Muttoni A, Ruiz MF. Shear strength of members without transversereinforcement as function of critical shear crack width. ACI Struct J2008;105:163–72.
[42] Leonhardt F, Walther R. The Stuttgart shear tests, 1961; contributions to thetreatment of the problems of shear in reinforced concreteconstruction. London: Cement and Concrete Association; 1962.
[43] Iyengar KTSR, Rangan BV, Palaniswamy R. Some factors affecting the shearstrength of reinforced concrete beams. Ind Conc J 1988;42:499–505.
[44] Lantsoght EOL. Progress report: experiments on slabs in reinforced concrete -part II: analysis of results [Stevin Report 25.5-12-10]. The Netherlands: DelftUniversity of Technology; 2012. p. 289.
[45] Cope RJ, Rao PV, Edwards KR. Shear in skew reinforced concrete slab bridges –analytical and experimental studies – a report to the Department ofTransport. Liverpool (UK): University of Liverpool; 1983. p. 219.
[46] ACI Committee 318. Building code requirements for structural concrete (ACI318-14) and commentary. Farmington Hills (MI): American Concrete Institute;2014.
[47] Yang Y, den Uijl JA, Walraven J. Shear capacity of reinforced concrete beamsunder complex loading conditions. In: The 9th FIB International PhDsymposium in civil engineering. Karlsruhe: Karlsruhe Institude ofTechnology; 2012. p. 43–8.
[48] Kani MW, Huggins MW, Wittkopp RR. Kani on shear in reinforced concrete.Canada; 1979.
[49] Lantsoght EOL. Shear in reinforced concrete slabs under concentrated loadsclose to supports. Delft University of Technology. PhD Thesis; 2013.
[50] Haritos N, Hira A, Mendis P, Heywood R, Giufre A. Load testing to collapse limitstate of Barr Creek Bridge. Transp Res Rec 2000:A92–A102.
[51] Azizinamini A, Boothby TE, Shekar Y, Barnhill G. Old concrete slab bridges. 1.Experimental investigation. J Struct Eng-ASCE 1994;120:3284–304.
[52] Azizinamini A, Shekar Y, Boothby TE, Barnhill G. Old concrete slab bridges. 2:analysis. J Struct Eng-ASCE 1994;120:3305–19.
[53] Miller RA, Aktan AE, Shahrooz BM. Destructive testing of decommissionedconcrete slab bridge. J Struct Eng-ASCE 1994;120:2176–98.
[54] Aktan AE, Zwick M, Miller R, Shahrooz B. Nondestructive and destructivetesting of decommissioned reinforced concrete slab highway bridge andassociated analytical studies. Transport Res Rec: J Transport Res Board1992;1371:142–53.
[55] Jorgenson JL, Larson W. Field testing of a reinforced concrete highway bridgeto collapse. Transport Res Rec: J Transport Res Board 1976;607:66–71.
[56] Bagge N, Shu J, Plos M, Elfgren L. Punching capacity of a reinforced concretebridge deck slab loaded to failure. IABSE 20152015.
[57] Bagge N, Sas G, Nilimaa J, Blanksvard T, Elfgren L, Tu Y, et al. Loading to failureof a 55 year old prestressed concrete bridge. Helsinki (Finland): IABSEWorkshop; 2015. p. 130–7.