and 50 mm diameter. It is adjusted to be collinearly impacted by the striking bar shot out from the air gun. Two sets of semiconductor strain gages, a and b , are cemented diametrically at a distance of 6d and 12d from the contact end the input bar with a specimen bar for measuring incident stress waves into the specimen bar [11]. One end of the concrete specimen bar is arranged in tight contact with one end of the input bar, while the other end of the specimen bar is released from stresses. The factors of reflection and transmission at the interface between the input bar and the specimen bar can be calculated using both the material properties of the input bar and the concrete specimen. In the present combination, α = - 0.25 and β = 0.75, where 1+ α = β on the assumption that all the incident stress, reflection stress and transmitted stress are taken to be compressive [2]. All specimens are equipped with two strain gages pasted diametrically at two locations 1 and 2, respectively, to measure directly stress waves propagated in a specimen bar. In order to specify the tensile break time, four crack gages (KYOWA, KV-5C) are also mounted at the center position of the specimens. The response signals trapped at those locations are passed through bridge boxes to a four-channel digital oscilloscope (Nicolet, Model 400). 500 βσ βσ βσ [mm] 750 1500 Strain gage a b C 100 50 β α σ Specimen bar Crack gage 1 2 Gage 2 βσ Gage 1 Tensile wave Compressive wave Striking bar Input bar Time [µsec] Figure 2: Experimental arrangement and superposition of tensile stress waves in concrete specimen. IMPACT TENSILE STRENGTH AND STRAIN RATES Cumulative Fracture Probability The impact tensile strength is determined from reading off the intensity of the tensile stress waves measured by the strain gages cemented on the concrete specimen bar. Suitable statistical analyses may be required to treat the dispersion of the experimental data in relation to the concrete strengths. A Weibull distribution was applied to not only the impact tensile strength but also the static strengths of the concrete specimen, as shown in Figure 3. To plot the i-th ranked sample from a total of n number of fractured specimens, a median-rank position was adopted, which is the distribution function Fi expressed approximately in terms of Fi = (i-0.3)/(n+0.4). The data plotted on the Weibull probability paper, i.e., lnln[1/(1-Fi)] versus ln( σ) lay on straight lines. The regression lines were drawn by means of the method of least squares. The Weibull modulus (shape parameter) m of each plot and the scale parameter ξ (fracture probability 63.2%) can be found from the Weibull distribution. The mean stress µ and the standard deviation s.d. can be calculated based on such data. The statistical results of the concrete used for this study are shown in Table 3. Comparing the impact tensile strength with the static tensile strength, it is worth noting that the tensile strength of the concrete is significantly influenced and increased by loading rates. Here the impact tensile strength, σit , in Table 3 denotes the minimum tensile stress to break the concrete specimen under impact loading in this experimental method. The impact splitting-tension test was also performed, but the detail is omitted.
RkJQdWJsaXNoZXIy MjM0NDE=