Experimental and Simulation Study of Torsional Ultrasonic Waveguide Sensors for Liquid Level Measurement

Accurate liquid level measurements in BWR and PWR feedwater systems are challenging due to harsh radiation, high temperature, and pressure environments, limiting available instrumentation to simple analog devices and hindering efficiency improvements. This study presents experimental and simulation results for torsional ultrasonic waveguide sensors designed for liquid level measurement. The sensors, fabricated with radiation-resilient materials, include a waveguide with magnetostrictive strips and a drive coil within a biasing coil. This technology measures fluid levels by launching and receiving torsional elastic waves along a solid waveguide. When the waveguide is coupled to the surrounding fluid, the propagation speed of torsional stress waves in a rod with a non-circular cross-section decreases, allowing fluid level determination through propagation delay time. Finite element analysis (FEA) simulations were employed to perform design iterations and optimizations to maximize measurement sensitivity for given geometries and operating conditions. In this study, two 1-meter-long steel waveguides with different rectangular cross-sections (6×4 and 6×2 mm) were selected. Laboratory experiments involved immersing the waveguides at various water levels, emitting and receiving torsional elastic waves at the top end, and recording data with a digital oscilloscope. Results showed that the propagation time delays of the torsional waves in both waveguides were linearly correlated with water level changes, with the waveguide having a 6×2 mm cross-section demonstrating higher sensitivity. The experimental procedures were simulated in ABAQUS, and the simulation results showed good agreement with the experimental data for both waveguides. The experimental and simulation analyses indicate the effectiveness of utilizing torsional waves for fluid level measurement. The strong agreement between experimental and simulation results validates the FEA models, supporting further design iterations.