Comparison of measured and finite element-based diffraction correction (FEDC) for ultrasonic measurement systems in a gaseous medium.

Diffraction corrections are essential in many ultrasonic systems utilizing one-way, two-way, and three-way measurements, or a combination thereof. Recent works have indicated that high-precision measurements may call for a diffraction correction model accounting for a full set of electrical and mechanical boundary conditions at transmission, reflection, and reception, such as the finite element diffraction correction (FEDC) model. However, the FEDC model has yet to be verified experimentally. In this work, the frequency-domain diffraction correction in an ultrasonic measurement system is measured, and compared to the FEDC model and a commonly used piston-type diffraction correction. One-way and three-way propagation are studied for an example system consisting of two piezoelectric ceramic disks in air separated by 0.10 and 0.15 m, vibrating in the frequency range of their lower radial resonances (50-300 kHz). Comparisons with measurements indicate that the FEDC model generally provides improved descriptions of diffraction effects compared to piston-type models. Specifically, mean absolute errors are decreased by 0.4-4.6 dB and 37-160° when using the FEDC model in the studied example, for magnitudes and phase angles, respectively. This shows that use of the FEDC model to correct for diffraction effects may increase the accuracy of high-precision measurement systems.