Aimed at the absolute calibration of infrasound sensors at very low frequencies, an upgraded and improved infrasonic pistonphone has been developed. The pistonphone was designed such that a very narrow clearance between the piston and its guide was realized based on an automatically-centered clearance-sealing structure, and a large volume rigid-walled chamber was also adopted, which improved the leakage time-constant of the chamber. A composite feedback control system was applied to the electromagnetic vibrator to control the precise motion of the piston. Performance tests and uncertainty analysis show that the leakage time-constant is so large, and the distortion of the sound pressure is so small, that the pistonphone can be used as a standard infrasound source in the frequency range from 0.001 Hz to 20 Hz. The low frequency property of the pistonphone has been verified through calibrating low frequency microphones. Comparison tests with the reciprocity method have shown that the pressure sensitivities from the pistonphone are not only reliable at common frequencies but also have smaller uncertainties at low frequencies.
Laser heterodyne interferometry plays a crucial role in measuring the velocity of water particles during the calibration of hydrophones with the optical method. The velocity of water particles acts as an indicator of acoustic-pressure variations and can be used to evaluate the stability of the acoustic field. The calibration of hydrophones requires a stable acoustic field environment; currently, though, the assessment of acoustic field stability is largely subjective. This study introduces the Normalized Dynamic Time-Warping (NDTW) algorithm, which objectively evaluates acoustic field stability. Sine-fitting is applied to the region of interest in the measured signal to obtain a reference signal. Subsequently, the NDTW algorithm is used to calculate the difference between the measured and reference signals, enabling the assessment of acoustic field stability. The NDTW algorithm effectively identifies subtle differences between signals and addresses the accumulation errors arising from varying signal lengths. The calibration results showed that for signals of high quality within the identified frequency band, the calibration outcomes obtained using the NDTW algorithm deviated from the reciprocity method by no more than 0.7 dB. For frequency bands with poor signal quality identified by the NDTW algorithm, the deviation between the calibration results and the reciprocity method exceeded 0.7 dB.