Development of Ultrasonic Devices for Non-destructive Testing: Ultrasonic Vibro-tactile Sensor and FPGA-Based Research Platform

Abstract

This thesis is focused on the development of ultrasonic devices for industrial non-destructive testing (NDT). Ultrasound is generated from mechanical vibrations and then propagates through the medium. Ultrasonic devices can make use of the ultrasound in both aspects, vibrations and propagations, to perform inspections of the objects. To this end, two devices were developed in this research, each pertaining to NDT of the objects. The first device is the vibro-tactile sensor which aims to estimate the elastic modules of soft materials with minimally invasive technique. Inspired by load sensitivity studies in the high-power ultrasonic applications, vibration characteristics in resonance were utilized to perform the inspection. Only a minimal force to ensure contact with the object surface needs to be applied for a vibro-tactile sensor to perform inspection of the object; hence, it can be used for in-vivo measurement of the soft materials’ elastic moduli without causing severe surface deformation. The design and analysis of the device were carried out using the electro-mechanical analogy to address the electro-mechanical nature of piezoelectric devices. The designed vibro-tactile sensor resonates at ~40 kHz and can be applied to differentiate the elastic modulus of isotropic soft samples with a range from 10 kPa to 70 kPa. The second device developed is a field-programmable development platform for ultrasonic pulse-echo testing. Ultrasonic testing, utilizing sound wave propagation, is a widely used technique in the industry. The commercially available equipment for industrial NDT is highly dependent on the competence of the inspector and rarely provides the access to raw data. For successful transition from traditional labor-intensive manufacturing to the next generation “smart factory” where intelligent machines replace human labor, inspection equipment with automated in-line data collection and processing capability is highly needed. To this end, a flexible platform which provides the access to raw data for algorithm development and implementation should be established. Therefore, an affordable, versatile, and researcher-friendly development platform based on field-programmable gate array (FPGA) was developed in the research. Both hardware and software development tools and procedures were discussed. In the lab experiment, the developed prototype exhibited its competence in NDT applications and successfully carried out hardware-based auto-detection algorithm for mm-level defects on steel and aluminum specimens. Comparisons with commercial systems were provided to guide future development.