170 H. Nozato et al. Fig. 18.1 Principles of shock calibration Fig. 18.2 Shock calibration system the displacement by the shock. With respect to the demodulation process from the quadrature signal to the acceleration waveform, we implemented phase unwrapping and differentiation twice through a Butterworth digital low-pass filter with a cut-off frequency of 5 kHz [5]. The accelerometer output is also passed through the Butterworth digital low-pass filter with a cut-off frequency of 5 kHz. 18.3 Experimental Results Figure 18.3a–c and d–f show typical experimental waveforms for two cases with accelerations of 50 and 10,000 m/s2, respectively. Here Fig. 18.3a, d show the acceleration measured by the He-Ne laser interferometer, and Fig. 18.3b, e show the voltage output from the combined piezoelectric accelerometer and charge amplifier. Figure 18.3c, f show expanded graphs around zero voltage for Fig. 18.3b, e, respectively. ISO 16063-13 for shock calibration defines the shock sensitivity of an accelerometer as two peak ratios between the input acceleration and the accelerometer output, as shown in Eq. (18.1). SV D VP AP (18.1) where: SV is the shock sensitivity, Ap is the peak value of the acceleration input to the accelerometer, Vp is the peak value of the accelerometer output. However, Vp includes the frequency response effect of not only the piezoelectric accelerometer but also the charge amplifier. Thus, by eliminating the effect of the charge amplifier, the charge sensitivity of the piezoelectric accelerometer can be calibrated. 18.4 Charge Amplifier and Virtual Amplifier Figure 18.4 shows the frequency response of a Brüel & Kjaer (BK) 2635 charge amplifier which has high-pass characteristics in the low-frequency region. Due to the effect of these high-pass characteristics, the charge amplifier output voltage with zero shift results in a zero voltage drift, which results in a large error component in the shock calibration. To reduce the effect of the
RkJQdWJsaXNoZXIy MTMzNzEzMQ==