Deep Ultrasound Enhancements Final Report

This study involves collaboration between Los Alamos National Laboratory and Lawrence Livermore National Laboratory to enhance and optimize LANL's ultrasonic inspection capabilities for production. Deep-penetrating ultrasonic testing enhancement studies will extend the current capabilities, which only look for disbonds. Current ultrasonic methods in production use 15-20 MHz to inspect for disbonds. The enhanced capabilities use 5 MHz to penetrate to the back surface and image the back surface for any flaws. The enhanced capabilities for back surface inspection use transducers and squirter modifications that can be incorporated into the existing production system. In a production setup the current 15-20 MHz transducer and squirter would perform a bond inspection, followed by a deep inspection that would be performed by simply swapping out the 5 MHz transducer and squirter. Surrogate samples were manufactured of beryllium and bismuth to perform the ultrasonic enhancement studies. The samples were used to simulate flaws on the back surface and study ultrasound's ability to image them. The ultrasonic technique was optimized by performing experiments with these samples and analyzing transducer performance in detecting flaws in the surrogate. Beam patterns were also studied experimentally using a steel ball reflector to measure beam patterns, focal points, and sensitivities to better understand the relationship between design and performance. Many transducers were evaluated including transducers from LANL's production system, LLNL, and other commercially available transducers. Squirter design was also analyzed while performing experiments Flat-bottom holes and ball-mill defects of various sizes were introduced into the samples for experimentation. Flaws depths were varied from .020'' to 0.060'', and diameters varied from 0.0625'' to 0.187''. The smallest defect, .020'' depth and 0.0625'', was detected. Ultrasonic amplitude features produced better images than time-of-flight features. 5 MHz was the optimal frequency, because it was able to penetrate deepest into the materials. Squirter standoff distance was found to be significant, and a standoff of an inch or so with an opening that is larger than the diameter of the transducer is recommended. Recommendations for enhancement are made based on the results of the experimental studies on Be-Bi samples. Future work should focus on deployment of a deep ultrasound technology and obtain initial results with real units. Efforts should be made to develop a reference sample for calibration. A reference sample would optimize reliability and sizing for test under production environment conditions