Practical parameters for extending the velocity range when applying a partial inversion pulse to the ultrasonic time-domain cross-correlation method

Abstract This paper describes a method to reduce false velocity detection by applying a partial inversion pulse (PIP) to the ultrasonic time-domain cross-correlation (UTDC) method. False detection occurs when we attempt to expand the velocity range of UTDC (same as the aliasing limit of the pulsed Doppler method). Specifically, the effects of three parameters that are important for practical use and have not been discussed before were examined, and recommended values for them were determined. The UTDC method can obtain the flow velocity profile along an ultrasonic beam by transmitting and receiving pulses. The PIP is a waveform that is phase-inverted within a pulse that is half as short as a conventional waveform, and it can reduce false detection by simply applying to the UTDC method. The three important parameters are the width of the reference window, the number density of reflective particles in the measurement volume, and the threshold for the cross-correlation coefficient. The PIP waveform was modeled, and the influences of these parameters were evaluated by simulations that calculate the correlation coefficient for single pulses and for waveforms overlapped with up to three pulses. In addition, the method was also experimentally verified in a pipe flow. Tests were also carried out with various number densities of reflectors. From these results, the effect of applying the PIP was confirmed for all values of these parameters. Specifically, the PIP can shift and reduce the peak of the first sidelobe in the correlation coefficient. As recommended values for each parameter, the experimental results showed that the width of the reference window should be equal to the pulse width, the number density of reflectors should be about one in the measurement volume, and the threshold value of the correlation coefficient should be about 0.5–0.7.