PSI - Issue 28

A.M. Bragov et al. / Procedia Structural Integrity 28 (2020) 2174–2180 Author name / Structural Integrity Procedia 00 (2019) 000–000

2178

5

3. Test results Fine-grained concrete was subjected to compressive tests using the split Hopkinson pressure bar. Test samples were cylindrical with a base diameter of 18 mm and a height of 10 mm. The setup consisted of a pneumatic loading device - a gas gun with a control system a complex of measuring and recording equipment and a replacement set of measuring bars with a diameter of 20 mm made of aluminium alloy and steel. Registration of initial experimental data was carried out using strain gauges glued on the lateral surface of measuring bars signals from which were transferred to a digital storage oscilloscope, using schemes of dynamic tensometry. Then the oscillograms were saved digitally and processed using the original software. Figure 2 shows the graphs P 1 (t) , P 2 (t) , P average (t) and R(t) obtained after classical processing of test results. It can be seen that the force acting on the sample from the side of the incident bar does not coincide well with the force acting on the sample from the side of the transmitted bar. This fact does not allow drawing a conclusion about the state of equilibrium deformation of the sample and casts doubt on the correctness of determining the mechanical properties of the test brittle material.

Fig. 2. The graphs P 1 (t) , P 2 (t) , P average (t) and R(t) obtained after classical processing of test results.

The reason for these phenomena may be the dispersion of the waves, due to which the pulses recorded at a certain distance from the sample have a distorted shape. In processing the experimental data, procedure of dispersion shift of pulses in the measuring bars (Bragov et al. (2019)) was used. Figure 3 shows the graphs P 1 (t) , P 2 (t) , P average (t) and R(t) obtained after applying a dispersion correction. It can be seen that the force acting on the sample from the side of the incident bar coincides very well with the force acting on the sample from the side of the transmitted bar and the maximum value P average (t) practically did not change. Therefore, the unsatisfactory fulfillment of the main assumption of the split Hopkinson pressure bar method in the case of classical processing of experimental data, which is visible on the basis of the analysis of the forces acting on the sample from the measuring bars, does not much affect the value of the fracture stress of the material tested, which proportional to maximum value P average (t) . A pulse shaper made of a material with a low yield strength and placed on the impacted end of the incident bar use for the production of a non-scattering smooth incident pulse with a soft leading edge (Frew et al. (2002)). Figure 4 shows the graphs P 1 (t) , P 2 (t) , P average (t) and R(t) obtained in the experiment using a copper pulse shaper. It can be seen that the force acting on the sample from the side of the incident bar practically coincides with the force acting on the sample from the side of the transmitted bar, and the time histories of these forces are more sloping. Thus, the use of the pulse shaper leads to a decrease in the maximum amplitude of the incident wave and a decrease in the steepness of the material deformation diagram in the stress-time axes in the experiment with the pulse shaper as compared to the experiment in which the pulse shaper was not used at the same striker velocities.