The name Acousto-Ultrasonics (AU) is derived from a combination of Acoustic-emission monitoring and Ultrasonic characterization. "Invented" by Alex Vary at NASA-Lewis, the method consists of monitoring and analyzing - the acoustic signals received from a controlled and complete insonification of the test material.
Acousto-Ultrasonics is a highly sophisticated and advanced technique using digital signal processing and pattern recognition algorithms. As such it is orders of magnitude superior to any other conventional ultrasonic technique, where the human recognizance capability is much inferior to modern data processing, or where automatic scanning systems are limited by considering only the amplitude in a small portion (gate) of the ultrasonic signal. Acousto-Ultrasoncs considers the entire ultrasonic response, in time as well as in frequency, of the entirely insonified material.
One major task of setting up an acousto-ultrasonic system consists of the training or learning process. Several identical standard specimens are needed for each feature that the system will have to be able to extract. A number of measurements have to be repeated several times on each specimen. After many digital feature extraction processes on all the waveform data, templates are built for each feature.
The feature templates are tested by evaluating the same standard specimen and the system is tested for its feature regognition reliability. If the reliability is insufficient, the ultrasonic interrogation system (transducer frequency, number, and location) may have to be altered and the training process will have to be repeated.
The inspection system concists of a cradle with all the transducers and of a mechanical subsystem to move the parts and to place the transducers in contact with the parts. It further contains an ultrasonic pulser/receiver unit, a switch matrix to control all the transducers, DSP boards, and a Computer.
Automated Test Procedure
(1) After the part is in place, all transducers are brought into contact.
(2) Waveform data are being acquired, controlled by the switch matrix.
(3) The acquired waveform data are being processed in the following phases:
where each feature group is sequentially evaluated against feature group templates from the learning set, and decision about "good" or "bad"
A-Scans and Frequency Spectra
Corrosion of Riveted Plates
Ultrasonic eneregy plotted versus time and frequency
Acousto-Ultrasonics is usefull for the inspection of components that have repeatable shapes, i.e. for
The method has been used for
Vary, A., The Acousto-Ultrasonic Approach, in Acousto-Ultrasonics, Theory and Applications, J. C. Duke, Jr., ed., Plenum Press, 1988.
Srivastav, V. K., Prakash, R., Acousto-Ultrasonic Evaluation of the Strength of Composite Material Adhesive Joints, Acousto-Ultrasonics, Theory and Applications, J. C. Duke, Jr., ed., Plenum Press, 1988.
Vary, A., Acousto-Ultrasonics, in Vol. 2 of Non-Destructive Testing of Fibre-Reinforced Plastics Composites, J. Summerscales, ed., Elsevier Science Publishers, 1990.
Beall, F. C., Overview of Acousto-Ultrasonics Applied to Wood and Wood-based Materials, Proceedings of the Second International Conference on Acousto-Ultrasonics, Alex Vary, ed., Copyright ASNT, 1993.
Bartos, A.L., J.O. Strycek, R.J Gewalt, H. Loertscher, and T.C. Chang, "Ultrasonic and Acousto-Ultrasonic Inspection and Characterization of Titanium Alloy Structures," Nondestructive Characterization of Materials VIII, R. E. Green, Jr., et al., eds., Plenum Press, 1998.Shiloh, K., A.L Bartos, A. Frain, and E. Lindgren, "Ultrasonic Detection of Corrosion between riveted Plates", Proceedings of SPIE 3994-09, 2000.