Shear Strength of Prestressed Concrete T-Beams with Steel Fibers Over Partial/Full Depth

ACI Structural Journal, May/Jun 2006 by Thomas, Job, Ramaswamy, Ananth

The test results of nine shear-critical partially prestressed concrete flanged beams with and without steel fibers are presented in this paper. T-beam specimens were cast with three grades of concrete: normal strength (35 MPa [5.07 ksi]), moderately high strength (65 MPa [9.42 ksi]), and high strength (85 MPa [12.32 ksi]). For each grade of concrete, three beams were cast: a control beam without fiber reinforcement, a beam with fiber reinforcement over the full depth of the cross section, and a beam with fiber reinforcement only in the web portion. Test results indicated that the provision of fiber reinforcement only in the web portion appreciably improved the shear-resisting capacity of the partially prestressed) beams. A model to predict the shear strength of prestressed and reinforced (nonprestressed) concrete beams has been proposed. The proposed model is expected to predict the test results of reinforced concrete beams having steel fibers over partial and full depth.

Keywords: fibers; prestress; shear; strength.

(ProQuest-CSA LLC: ... denotes formulae omitted.)

INTRODUCTION

The assessment of the contribution of steel fibers to various shear-resisting mechanisms through the enhanced concrete cracking (tensile) strength, fiber pullout, and fiber dowel resistance at cracks has been investigated during the past three decades.1,2 The development and use of highstrength concrete (HSC) ( f'^sub cu^ > 80 MPa [11.60 ksi] where f'^sub cy^ [asymptotically =] f'^sub cu^ /1.2), which is relatively more brittle compared with normal-strength concrete (NSC) ( f'^sub cu^ > 30 MPa [4.35 ksi]) and moderately high-strength concrete ( f'^sub cu^ > 60 MPa [8.7 ksi]), may require the use of fiber reinforcement to mitigate the catastrophic shear failures in these beams. Thus, a need is apparent for fine-tuning existing rational approaches for the shear strength prediction developed based on the experimental results of NSC and moderately high-strength concrete. In the present study, the effect of concrete strength on the shear contribution due to steel fibers in reinforced concrete beams has been integrated with the prediction models for shear strength found in the literature.

The effects of concrete strength, prestressing force, shear span to depth ratio, amount of longitudinal steel, and width of the web on the shear strength of the prestressed concrete beams have been commented in many codes of practice.3-5 Many studies6-11 investigated the potential use of fibers as shear reinforcement in prestressed concrete beams. Many authors12-17 reported that the addition of steel fibers resulted in significant improvement of the shear capacity for reinforced (nonprestressed) beams. In the work of Cho and Kim,12 an upper-bound solution for shear resistance of fibrous concrete beams was obtained under a plane stress state by considering the equilibrium conditions on an assumed hyperbolic yield line in the shear span of the beam. The effect of the addition of fibers over the partial depth of prestressed concrete beams, however, has not yet been extensively investigated. An earlier study by Swamy and Bahia18 on using steel fibers over the partial depth of approximately 0.36h in the tension zone did not show reduced shear failures in a nonprestressed beam. In the present study, shear-critical prestressed concrete T-beams without and with fiber reinforcement over the partial (0.71h) or full depth were cast and tested to estimate the influence of fibers on the shear-resisting mechanisms.

RESEARCH SIGNIFICANCE

This paper provides test data pertaining to the shear strength of partially prestressed T-beams with steel fibers over their partial depth. The change in the mode of shear failure exhibited in the prestressed beams with moderately high-strength (65 MPa [9.42 ksi]) and high-strength (85 MPa [12.32 ksi]) grade concrete has been reported. An empirical model for the prediction of the shear strength of reinforced concrete and prestressed concrete derived from the regression analysis of 518 test data from this and other studies found in literature is presented. The presented model accounts for the contribution of fiber pullout mechanism across the shear crack expressed as a function of the matrix strength. The presence of fibers over the partial depth of the beam has been accounted for in the proposed shear strength model by introducing a partial depth fiber parameter. The shear strength contribution of the fibers has also been expressed in terms of equivalent stirrup spacing in prestressed concrete beams with concrete having no fibers.

EXPERIMENTAL INVESTIGATION

A total of nine T-beams were fabricated and tested in this study. All beams were 3.85 m (151.57 in.) long and were tested over a clear span of 3.6 m (141.73 in.). The longitudinal and cross-sectional details of the test specimens are shown in Fig. 1. The amount of longitudinal and transverse reinforcement was kept constant for all specimens. As all structural beams require nominal links to avoid problems resulting from marginal and unforeseen increases in the applied shear, stirrups were provided in all the test beams based on the governing criteria of minimum spacing (0.75d for the present case) specified by the Indian code of practice.5 The shear span to depth ratio was maintained constant for the entire test program (a/d = 2.65). The parameters that affect the shear strength of beams identified for the present study were the grade of concrete and the presence of steel fibers over partial or full depth. Pretensioned T-beam specimens were cast with NSC, moderately high-strength concrete ( f'^sub cu^ = 65 MPa [9.42 ksi]), or HSC. Three control beam specimens (one for each of the different concrete strengths with no fiber reinforcement), three T-beam specimens having fiber reinforcement over their full depth, and another three T-beams having fiber reinforcement only in the web portion, were cast to study the effect of the various parameters identified.