Ten-Year Study on Steel Fiber-Reinforced Concrete Beams Under Sustained Loads

ACI Structural Journal, May/Jun 2005 by Tan, Kiang Hwee, Saha, Mithun Kumar

Nine reinforced concrete beams with discrete steel fiber contents ranging from 0 to 2% were subjected to sustained flexural loading between 0.35 to 0.8 times the flexural capacity over a period of 10 years. The long-term deflections and maximum crack widths were found to decrease with increasing steel fiber content, with a reduction of 34 and 58%, respectively, at the end of the 10-year period in the case of the 2% fiber content. The deflections and crack widths were smaller for lower sustained loads. The modified ACI approach and effective modulus method predict the long-term deflections very well, while the adjusted effective modulus method slightly underestimates the same. Under four-point loading, the beams showed a reduced stiffness in the initial stage. After the previously sustained load level was attained, however, the beams exhibited essentially the same load-deflection characteristics as the corresponding ones that were not subjected to sustained loadings.

Keywords: beam; cracking; deflection; stiffness; strength.

(ProQuest Information and Learning: ... denotes formulae omitted.)

INTRODUCTION

Although the behavior of steel fiber-reinforced concrete (SFRC) beams in bending, shear, and torsion is well established (ACI Committee 544 1988; Balaguru and Shah 1992), their application in the field is still relatively narrow. This is partly due to insufficient information on the long-term performance of such members, particularly with respect to serviceability.

Steel fibers have been known to aid in deflection control as reported by Swamy and Al-Noori (1975) in a study on the flexural behavior of SFRC beams under four-point loading. In another study on beams under midpoint loading, Swamy, Al-Ta'an, and Ali (1979) further concluded that the presence of steel fibers led to an increase in the stiffness of beams, resulting in a substantial reduction of deflections. Kormeling, Reinhardt, and Shah (1980) tested SFRC beams with different volumes of longitudinal reinforcement and with different types of steel fibers. They also concluded that the presence of steel fibers led to a smaller beam deflection. To predict the instantaneous deflection of SFRC beams, various analytical models have been proposed. Craig et al. (1987) proposed a method to find the curvature of a section based on strain compatibility and equilibrium of forces and computed the deflection using the conjugate beam method. Soroushian and Reklaoui (1989) also proposed a similar method in which they employed an idealized stress-strain relationship for steel-fiber concrete.

Several studies have been conducted on the creep and shrinkage behavior of steel-fiber concrete, while studies on long-term behavior of SFRC beams under sustained loads are uncommon. Malmberg and Skarendahl (1978), using specimens measuring 40 × 40 × 160 mm, reported that steel fiber concrete with a fiber content of up to 2% underwent less shrinkage than plain concrete. Another study by Swamy and Theodorakopoulos (1979), using 500 × 100 × 25 mm plate specimens, revealed that inclusion of 1% fiber resulted in improved creep properties of concrete under flexure. On the other hand, compressive creep tests conducted by Houde, Prezeau, and Roux (1987), using 150 x 300 mm cylindrical concrete specimens with 1% fiber content, and by Balaguru and Ramakrishnan (1988), using the same type of cylindrical specimens but with 0.5% fiber content, showed an increase in creep strains compared with the plain concrete specimens.

In this paper, the behavior of SFRC beams with different fiber contents under different sustained loads for a period of 10 years is reported. A comparison between the observed deflections and analytical predictions is made. Both linear and nonlinear empirical formulas are proposed to predict the long-term (incremental) crack widths from the instantaneous crack widths under the same load. Also, the beams were tested in flexure under four-point (third-point) loading after the sustained loads were removed and the results are compared with those of the original beams tested by Tan, Paramasivam, and Tan (1994a,b) that were not subjected to sustained loads.

RESEARCH SIGNIFICANCE

This paper reports the long-term deformation and cracking characteristics of reinforced concrete beams with discrete steel fibers under sustained loads for a period of 10 years. The steel fiber content was varied from 0 to 2%, while the sustained load level was varied from 0.35 to 0.8 times the flexural strength of the beam without steel fibers. The steel fibers were found to effectively control the long-term deflections and crack widths. The modified ACI approach and effective modulus method (EMM) were found to predict the long-term deflections very well. Empirical formulas for the prediction of long-term crack widths are also presented.

EXPERIMENTAL INVESTIGATION

Nine beams, as indicated in Table 1, with varying steel fiber contents that were previously reported (Tan, Paramasivam, and Tan 1994a,b) had since been subjected to various levels of sustained loads for a period of 10 years. Each beam, measuring 100 × 125 mm in cross section, was simply supported over a clear-span length of 1800 mm. The beam specimens were fabricated using normal portland cement, natural sand, crushed granite of 10 mm maximum size, and water, mixed in the proportion of 1:1.5:2.5:0.5 by weight to achieve a concrete cube compressive strength of 40 MPa. Hooked-end steel fibers, 30 mm in length and 0.5 mm in diameter, were used as fiber reinforcement, while two T10 longitudinal steel bars (10 mm in diameter) were used as tension reinforcement and two T6 bars (6 mm in diameter) as compression reinforcement at a distance of 99 and 24 mm from the top face, respectively, as shown in Fig. 1(a). R6 stirrups were also provided at a spacing of 75 mm along the beam length to prevent it from failing in shear.