Direct tension test 

Context in source publication

Context 1

... mm 460 mm for the wet-screened concrete specimens. Both sets were cast in the steel moulds. The length of 80 mm was cut from both ends of the specimens by disc cutting machine before the tests. Then, the ends were polished by grinding machine. Shapes and sizes of the specimens are shown in Figure 1 and Table 2 respectively. Sizes of the TPB specimens are: 1700 mm 400 mm 240 mm for the dam concrete specimens, and 900 mm × 200 mm × 120 mm for the wet-screened concrete specimens. The specimens were cast in wooden moulds and pre-formed crack was fabricated by the steel plate with thickness of 2 mm, which was coated with lubricating oil on both sides. The steel plate was pulled out from concrete within 3 hours after the initial setting. Sizes of the beams are shown in Table 3. In the table, B, D, L and a 0 are the width, depth, length and initial crack length of the TPB specimen, respectively. To obtain complete stress-deformation response in DT test, a very stiff testing machine and closed loop control machine is needed (Shah et al . 1995). Also proper grips and extreme care in alignment of load line of action are important in order to avoid eccentric loading. The DT tests were conducted by employing a very stiff servo-hydraulic closed-loop testing machine with stiffness of 6000 kN/mm and load capacity of 3000 kN. The DT test set-up is shown in Figure 2. Specimens were glued to the end platens of testing machine with epoxy. Four extensometers were installed around the specimen symmetrically for measuring the uniaxial tension deformation. The extensometers have a displacement range of ± 5 mm. The average displacement from the four extensometers was used to calculate tensile strain. The maximum displacement from the output of four extensometers was employed as the feedback signal to control the machine. The speed used for strain control is 4–6 με /min. The actual load-carrying capacity and stress distribution of each section of a specimen is different because the inhomogeneity which could result from the random distribution of coarse aggregates and the original flaws including initial cracks and voids. The cracks occurred first at the weakest section of the specimen. Note that eccentricity may not be avoided when a crack initiates from one side of the specimen. The issues of possible eccentricity and the end conditions have been dealt in details by others (Vervuurt et al . 1996; van Mier et al . 1996). The axial tensile failure sections of dam concrete specimens DT73, DT74 and DT75 were close to the middle part of the specimens as well as the wet-screened concrete specimen DT66. However, the failure section of wet-screened concrete specimens DT67 and DT69 located near the lower part of the specimens. The cracks of most specimens initiated from one side, then propagated toward the other side. However, the minority of specimens initiated from two sides, then propagated inward. The tests were conducted on a very stiff testing machine. A load cell with a capacity of 100 kN was adopted in the tests; the accuracy was ± 2% of the maximum applied load. The crack mouth opening displacement (CMOD) was measured by a displacement sensor whose capacity and accuracy are 5 mm and 0.5 μ m respectively. It was possible to obtain the complete load-CMOD curves (see Figure 3). The tested P CMOD curves of dam concrete and wet-screened concrete are shown in Figure 4. It is obvious that the maximum load P max of dam concrete is higher than that of wet-screened concrete. The measured stress-deformation curves ( curves) of dam concrete and wet-screened concrete by the DT tests are shown in Figure 5. From the curves, w curves were obtained by the following procedure and the sketch shown in Figure 6. The total deformation δ tested by the DT test can be expressed simply by ...