*3.3. Experimental Design*

In the experiment, the impact velocity of the bullet was altered by changing the driving air pressure so as to obtain the stress–strain curve of the material under different strain rates. A total of three kinds of loading speeds (impact speeds of 5 m/s, 10 m/s, and 15 m/s) for the SHPB impact compression experiments were carried out. Three repeated experiments were carried out for each speed state, and the average curve of these three experiments was taken as the stress–strain curve of the material under the strain rate. The experiment process of ultra-early-strength cement-based material is shown in Figure 3. The experiment loading speed and test specimen number are shown in Table 2.

**Figure 3.** Physical photos of the test process.

**Table 2.** Experimental program.


#### *3.4. Calibration of SHPB*

The basic principle of SHPB is to decouple the wave propagation effect and the strain rate effect of the material and then separate the strain rate effect of the material. Figure 4a shows the electrical signal curve collected by the data acquisition system, which represents the curve of voltage change over time. The recorded data can be restored to stress and strain curves by data processing software. Figure 4b is the three-wave diagram of the incident wave, reflected wave, and transmitted wave obtained from the experiment. The comparison wave is the incident wave + the reflected wave. It can be seen from the figure that the transmitted wave is in good agreement with the comparison wave, which meets the criterion of *εt*(t) = *εr*(t) + *εi*(t), proving that the test data are valid and can be used for analysis.

In order to make the stress pulse have enough time to reflect back and forth before the failure of the ultra-early-strength cement-based material specimen to obtain a uniform distribution of stress in the specimen, wave shapers were installed on the impacted end of the incident bar, which can eliminate the overshoot and wave oscillation of the stress wave caused by the dispersion effect of the large-size SHPB device, and it is helpful to obtain the true response characteristics of the material. In the experiment, under different loading conditions, different shapers were selected. Among them, the shaper used under low strain rate loading conditions (impact velocity 5 m/s) was a Φ16 × 0.5 mm copper sheet. Under the condition of medium strain rate loading (impact velocity 10 m/s), a Φ30 × 2 mm copper sheet shaper was used. Under high strain rate loading conditions (impact velocity 15 m/s), the shaper is a Φ30 × 4 mm copper sheet. By choosing a proper shaper, the dispersion effect can be eliminated effectively, the change in the waveform when the wave propagates in the waveguide bar is reduced, and the accuracy of the experiment is improved.

**Figure 4.** Original stress wave. (**a**) SHPB test collection wave. (**b**) The three waves graph.

.

Assuming that the strain rate is constant during the loading process, the average strain rate during loading can be determined by Formula (1):

$$
\dot{\varepsilon} = \varepsilon/t \tag{1}
$$

If *t* = 100 μs and the failure strain *ε* = 8500 με, it can be determined that the highest strain rate that can satisfy the stress uniformity is 85 s<sup>−</sup>1. It is particularly pointed out that the strain rate of the specimen is not constant during the experiment. During the data processing, the average strain rate of the loading stage before the failure of the specimen is taken as the average strain rate.

#### **4. Test Results**
