Search

Otto Buck was an excellent scientist and a close personal friend. I first met Otto at the newly founded North American Aviation Science Center which I joined in April, 1964, after leaving the Oak Ridge National Laboratory. Otto joined a week later from the Max Planck Institute in Stuttgart, Germany. We were assigned to work together, thus beginning a wonderful experience that lasted up to his passing in November, 1997. In our collaborations, Otto emphasized the theoretical modeling aspects of the work while I emphasized the experimental. The purpose of this paper is to recall in memoriam some of the highlights of Otto’ s early contributions to the scientific literature resulting from our collaboration. The period covered extends approximately from 1964 to 1974. Previous to this time, Otto had made significant contributions to several areas in metal physics including elasticity, radiation damage, and plastic deformation studies [1–8]. Because of the memorial nature of the paper, an attempt will be made to show highlights of Otto’ s developments only and not to emphasize the individual details and rigor that always characterized Otto’ s work. The reader is referred to the original publications to extract those.

The primary objective in calibrating ultrasonic transducers and measurement systems is to ensure the interchangeability of data. Thus, one wishes to create a situation in which experimental results, e.g. A-Scans or C-Scans taken at one time and place, can be directly compared in a quantitative sense to similar data taken at another time and place. The presently available tools are not sufficient to this task. As noted by Burley [1], there has been no single type of calibration standard that is suitable in all ultrasonic applications and inspection procedures. Instead, there is a wide range of reference standards that are used in different situations to calibrate transducers and measurement systems. Although these have served the community well in bounding the results which can be obtained by different systems, they do not ensure uniformity of results. A striking example for the case of eddy currents can be found in Fig. 4 of Ref. [2].

The Review of Progress in Quantitative NDE is the world's leading conference in reporting annually new research and development results in quantitative NDE. The conference reports on both fundamental and applied advances in NDE and promotes communication between the research and engineering communities and emphasize current reporting of work in progress. Attendees include representatives of academia (including students), industry, and government with approximately one-half coming from the United States and the other half from overseas. This volume represents the best report of ongoing work that is available anywhere. Connections and overlap with the medical diagnostic community are highlighted.

Noise-corrupted flaw signals with known contributions due to flaw scattering are needed to study flaw detection and characterization approaches. The inability to create these “known flaw” signals which are corrupted with acoustic noise (e.g., grain scattering noise) has presented a problem to the nondestructive evaluation (NDE) community for years. An approach is needed for creating noise-corrupted flaw signals which can be used as a basis for studying the impact of different flaw types, measurement system configurations, signal-to-noise ratios (SNRs), acoustic noise types, etc. on digital signal processing algorithms which are being developed.

Unified life-cycle engineering (ULCE) is an approach that combines aspects of concurrent engineering, total quality management, and retirement-for-cause (RFC) strategies at the early design stage of components [1–3]. In order to implement ULCE approaches, inspectability, reliability, and the other “ilities” of Fig. 1 must be linked to areas such as quality assurance, life-cycle costs, and materials and processes in a comprehensive design environment.

Thick composites are in use in critical applications and are proposed for still others. It is important to measure the elastic moduli of thick composites for two reasons: (1) design data on stiffness, and (2) prediction of feasible wave paths for ultrasonic waves for NDE. Except for one earlier report, (1) only relatively thin composites of relatively simple symmetries have been measured for their elastic moduli (2–6), Now, it is becoming necessary to measure thick composites of feasible engineering lay-ups. These often provide the complexity of orthorhombic symmetry locally in a specimen combined with curvature in the gross structure. In this work, specimens cut from thick structures will be treated in the same way as crystals to measure the elastic moduli by means of ultrasonic wave velocities. Results on three structures will be presented.

In ultrasonic nondestructive evaluation, experimental measurements of the scattered wave field resulting from sonification of a flaw are corrupted with acoustic noise. Acoustic noise results from non-flaw related scattering or reflection of the incident waves. In many probabilistic approaches to flaw detection, classification, and characterization, a stochastic model for a noise-corrupted flaw signal is utilized where acoustic noise is assumed to be an uncorrelated, Gaussian random variable with zero mean. In addition, it is assumed that an estimate of the average power spectra of the noise is available [1–3]. The goal of the work presented here was to measure and analyze acoustic noise as a random variable. Emphasis was placed on evaluating these assumptions and on estimating the average power spectra of the noise.

A novel technique has been developed for generating ultrasonic beams with spatial profiles of amplitude governed by a truncated Bessel function or Gaussian function [1,2]. Bessel beams have very unique properties; in optics Bessel beams have been shown to be diffractionless (J. Durnin et al, 1987 [3,4]). In a related work, R. W. Ziolkowski et al [5] reported experimental measurements of “acoustic directed energy pulse trains” generated by synthetic line array of ultrasonic transmitters in water. However, a Bessel function ultrasonic transducer has never been reported before. Gaussian beams also have desirable properties; they are very easy to model analytically, and a circular Gaussian function ultrasonic transducer is free of near-field nulls and far-field sidelobes associated with conventional “piston source” transducers [6]. At least three designs of Gaussian transducers have been reported in the literature in the past 30 years [7–9]. We report a method in which piezoelectric ceramic elements are poled with nonuniform electric fields shaped like Bessel or Gaussian functions such that the resulting polarization (and hence the ultrasonic amplitude) follows that of the applied poling field. Like conventional piston source transducers, such Bessel or Gaussian transducers also possess the simple “parallel plate capacitor” configuration and can be packaged likewise. Beam profiles and propagation behavior of these Bessel and Gaussian transducers have been measured experimentally in an immersion tank and the results compared well with model predictions

The use of prior information is an important component in the ultrasonic detection, classification, and characterization of flaws. In order to take full advantage of advanced digitally based approaches to flaw detection, classification, and characterization, use of prior information will be critical. Some advanced techniques involve probabilistic approaches which start with a stochastic model for a flaw signal in which the flaw’s scattering amplitude is assumed to be an uncorrelated, Gaussian random variable with zero mean and known variance [1–4]. The goal of the work presented here was to analyze scattering amplitude as a random variable with emphasis on evaluation of these assumptions.

Thick composites are in use in critical applications and are proposed for still others. It is important to measure the elastic moduli of thick composites for two reasons: (1) for design data on stiffness, and (2) for prediction of feasible wave paths for ultrasonic waves for NDE. Previously only relatively thin composites of relatively simple symmetries have been measured for their elastic moduli. Now, it is becoming necessary to measure thick composites of feasible engineering lay-ups. These generally provide the complexity of orthorhombic symmetry locally in a specimen combined with curvature in the gross structure. In this work, specimens cut from thick structures will be treated in the same way as crystals to measure the elastic moduli by means of ultrasonic wave velocities. Results on one structure will be presented. Difficulties will be analyzed.

In ultrasonic nondestructive evaluation (NDE) studies, impulse response is often used to evaluate internal defects. Under the far-field Born approximation [1], the impulse response can be written as the product of the second derivative of the area function with respect to the coordinate along the line-of-sight and a scattering constant which depends on the material properties as well as scattering directions [2]. The line-of-sight is a straight line along the illumination direction for pulse-echo tests, and the area function is an artificial time domain waveform equal to the target cross-sectional area intersected by an imaginary transverse plane travelling along the line-of-sight. Hence the double integration of the impulse response yields the product of the area function and the scattering constant. This product is also known as the ramp function response or the ramp response. Dividing the ramp response by its integral results the area function per volume of the target (normalized area function) [2] thus eliminating the unknown scattering constant. Since the area function (or its normalized form) contains the geometric information about the target, it has been employed in studies of both radar [3–5] and ultrasonic signal imaging [6]. In this paper, an ultrasonic CT algorithm is develped, using the area function from the Born approximation previously suggested by Tam [7]. A similar imaging technique for radar signals can be found in an earlier work done by Das and Boerner [5].

It has been shown in other papers in this volume (1,2) that the application of deconvolution, diffraction, and attenuation corrections to received ultrasonic signals is sufficient to extract the absolute value of the flaw scattering amplitude from the signal. This capability provides a new opportunity to obtain additional flaw characterization information from the ultrasonic signal. More specifically, it is postulated upon reasonable grounds that the absolute magnitude of the front surface ultrasonic echo from a flaw can be related to both the flaw dimensions and its acoustic impedance. Assuming that the size can be obtained by other means, the front surface echo can then be utilized to help identify the flaw. The purpose of this paper is thus twofold: 1) to utilize the diffraction and attenuation corrections developed by Thompson and Gray (3) to obtain absolute values of flaw impulse responses, and 2) to investigate the feasibility of using absolute values of front surface echoes to determine a flaw’s identity.

Work is currently underway to develop and test prototypic ultrasonic instrumentation that will produce unipolar stress pulses in a pulse-echo mode. The instrument will be used to detect and characterize flaws in materials and for the measurement of material properties important in material processing applications. In conjunction with this work, a Gauss-Hermite Beam Model [1,2,3,4] is used to predict the response of various transducer element geometries to maintain the unipolar response over a longer propagation distance.

To exploit theoretical advances in elastic wave inverse scattering, an automated multiviewing ultrasonic transducer system and the associated signal processing algorithms have been developed at the Ames Laboratory for the reconstruction of the size, shape, and orientation of volumetric flaws [1]. The flaw sizing algorithm is based on elastic wave inverse scattering theories in the long and intermediate wavelength, regime [2,3] and the three-dimensional reconstruction algorithm finds the equivalent ellipsoid that best fits the flaw sizes in the various viewing directions [4,5]. The original multiprobe system consists of six peripheral transducers equally spaced in a circle surrounding one transducer at the center. The peripheral transducers may be tilted at an angle toward the center to increase the aperture and can also be translated along their respective axes to allow an equilization of the acoustic propagation time. The axis of the aperture cone is normally placed perpendicular to the part surface. The flaw sizing procedure was a one-dimensional inverse Born algorithm to determine the flaw’s centroid-to-tangent plane distances for a number (normally 13 or 19) of pulse-echo or pitch-catch scattering directions within a finite aperture cone. The flaw sizes are then used as inputs to a nonlinear least squares regression program to yield a complete geometric reconstruction in the form of three semi-axes and three Euler angles of the best-fit ellipsoid. Using this system, successful reconstructions have been obtained for both oblate spheroidal (disk-like) and prolate spheroidal (rod-like) inclusions and voids. The readers are referred to a complete description of the system in Ref. 1.

On-line NDE practices typically emerge through a sequence of steps starting with the selection of a particular technique, proceeding with definition of configurations and procedures, and culminating in qualification through an experimental demonstration of performance. Because of the absence of a well-developed analytical foundation, this process has been, by necessity, highly empirical in nature. The theoretical advances in the understanding of field-flaw interactions which have occurred over the last decade provide the tools necessary to adopt a more analytical approach to the engineering of a new NDE system. This paper explores the present status of these tools and discusses different ways in which they have been, or might be, used to quantify and improve inspection reliability.

Advances in elastic wave scattering and inversion techniques have shown that advances in transducer technology are needed in order to fully exploit them. This is particularly true in the case of flaw sizing algorithms in which it has been demonstrated that a need exists for transducers with specified bandwidth characteristics. As a part of an effort to develop a composite, multiviewing ultrasonic transducer for flaw characterization, a computational tool has been developed which provides a convenient way to select driver pulse shapes and transducer characteristics which optimize this property. The purpose of this paper is to discuss this computational tool.

As a part of a project to develop an ultrasonic multiviewing transducer which is capable of providing sufficient information for flaw reconstruction, results have been obtained that address certain phases of the required signal processing routine. It is well known that individual ultrasonic transducers show a considerable variation in signal responses. Inasmuch as the multiviewing transducer uses a sparse array of individual transducers arranged in a particular geometry, it becomes important that signal processing routines be developed and applied which overcome these variabilities. There are, of course, several possible sources of variability in transducer performance. These include those of measurement (diffraction, attenuation and effects associated with imperfect deconvolution processes), and those due to fabrication (transducer design, materials, reproducibility of fabrication techniques, etc.). The results presented in this paper address only the former of these two. In particular, they represent a limited experimental assessment of the adequacy of specific data processing procedures for three different transducers and three different samples utilizing Weiner filter concepts and diffraction and attenuation corrections. Use is made of theoretical curves for scattering amplitude as an absolute standard of comparison and therefore as the criterion of adequacy.

The characterization of failure-initiating flaws in materials and components is an important task in the assurance of structural integrity. Based on ultrasonic signals scattered from the flaw in a number of directions, inverse scattering algorithms may be used to reconstruct the size, shape and orientation of the flaw. One of the more successful methods for elastic wave inverse scattering and flaw sizing has been the inverse Born algorithm in the long and intermediate wavelength regime.

The Wiener filter is currently used in the ultrasonic scattering amplitude estimation problem as a means of desensitization during deconvolution [1,2,3]. The work summarized here focuses on a Wiener filtering approach which incorporates a priori flaw and noise information. It will be shown that this approach leads to improved scattering amplitude estimates and improved radius estimates.

A number of alloys exhibit a phenomenon which has come to be known as ‘shape memory’. Without exception these alloys undergo a rather special type of martensitic transformation which results from one or another of the elastic constants1–7 weakening with changing temperature to the point that at the transformation temperature the transforming phase becomes both mechanically and thermodynamically unstable. ‘Shape memory’ is exhibited in the following way. A specimen at a temperature above the martensitic transformation is shaped to some desired form. The specimen is then cooled below its transformation temperature and deformed. Subsequent heating of the specimen through the transformation temperature causes reversion to the originally fabricated shape.

We have made a comparison of the ultrasonic attenuation coefficient and phase velocity obtained from two different experimental systems. The first set-up employed unipolar (either one relative maximum or minimum) ultrasonic pulses to interogate a specimen of porous woven graphite/epoxy composite. In the second, bipolar (one relative maximum and one relative minimum) ultrasonic pulses were used to interogate the same specimen. The replacement of bipolar pulses by unipolar pulses lead to an increase in fractional bandwidth of at least 200% for measurements of both phase velocity and attenuation coefficient. Most of this increase arose from improvements in sensitivity at lower frequencies. Gains of this nature will significantly improve the stability of flaw inversion algorithms in either thick composites or porous composites where attenuation losses prevent the acquisition of data at high frequencies. Use of unipolar pulses also doubles the temporal (spatial) resolution that may be obtained using ultrasound generated by any particular transducer; this is a consequence of the fact that the unipolar pulse will be one half as long as the bipolar pulse produced by the same transducer.

One of the major concerns in nondestructive evaluation (NDE) is the detection and characterization of cracks in structural materials. In the development of NDE methodologies, laboratory samples containing known flaws are very useful for the experimental verification of models and algorithms. Simulated flaws made to specific design serve to check the correctness of various aspects of an NDE technique under development. In addition, model flaws are also used in the calibration of nondestructive testing equipments. A familiar example is the ubiquitous flat bottom hole test blocks.

In this paper we report a new method for determining the size of flaws using the zeroes in the real and imaginary parts of the ultrasonic scattering amplitude [1]. It is observed that, for a wide class of flaws, the wavenumbers or frequencies at which the real and imaginary parts of the scattering amplitude go through zero are closely related to the flaw dimension in the scattering direction. This relationship remains remarkably stable for different interrogation directions and thus may serve as a basis for flaw sizing and reconstruction. In the decomposition of the scattering amplitude into its real and imaginary components, this method requires that the phase reference (or the “zero of time”) be placed at the centroid of the flaw. In many respects the method of flaw sizing using the zeroes is similar to the one-dimensional version of the inverse Born approximation (1-D IBA) [2], In fact, it employs the same input data as the 1-D IBA and consequently also shares certain common limitations such as the sensitivity of the sizing results to the zero of time determination and the available bandwidth of the scattering data [3,4]. However, flaw sizing using the zeroes is simpler and more straight forward because the sizes are obtained directly from the frequencies where the zeroes occur.

An automated multiviewing ultrasonic transducer and a first generation signal processing program have been developed for the purpose of detecting and characterizing flaws in materials. The multiviewing transducer has been designed and developed to exploit advances in theoretical inverse elastic wave scattering in the long and intermediate wavelength regime made in recent years. The signal processing algorithm has been assembled as a first step in the development of a decision-tree algorithm for flaw characterization. First results that have been obtained in a completely automatic mode are given in this paper. It is concluded that this new long and intermediate wavelength, model-based reconstruction is feasible and potentially very useful in quantitative NDE applications on real systems.

The purpose of the work described in this paper is the development of an ultrasonic measurement technique which provides a convenient way to detect dilute porosity conditions in materials and to extract certain properties of the flaw distribution which are important in failure prediction. Use has been made entirely of ultrasonic backscatter measurements; thus, the technique differs considerably from other investigations which lead to porosity determinations in that no reliance is placed upon either attenuation measurements or precise ultrasonic velocity measurements [1,2]. The technique thus possesses a distinct advantage for practical implementation, i.e., it is a “one-sided” measurement which does not require ultrasonic echo returns from an opposite face of the sample in order to be useful. At present, the work is limited to dilute porosity concentrations. Reasons for this limitation will become clear in the paper. With additional effort it is expected that this limitation can be removed and the work extended to larger concentrations.

Our keynote talk today has several objectives: First, we want to call attention, as some of us have before, to the interdisciplinary nature of NDE science and technology and some approaches for fostering RD we feel that they were both necessary and evolutionary. Many of you are probably familiar with these changes by now but there may be some residual concerns or questions in your minds. Since I was the initiating influence behind most of them, it is appropriate for you to hear me say what they are and what they aren’t, and to have an opportunity to question us. Finally and most importantly, we want to enlist your participation in the difficult task of identifying exploratory development programs—and we will try to define this term—which will benefit from the growing science base that all of us are helping to develop. We hope to stimulate the interaction between people such as you and the consumers of the evolving NDE technology in order to identify reduction-to-practice possibilities that should be pursued.

Considerable progress has been made in recent years in the development of signal processing algorithms for use in ultrasonic non-destructive evaluation which yield the size, shape, and orientation of a flaw. This kind of flaw information is necessary in order that failure predictions of materials and components can be made from non-destructive tests. The signal processing algorithms that have been developed for ultrasonics are based upon both direct and inverse approximate solutions to the elastic wave scattering problem, and cover various ranges of the parameter ka where \(k = \frac{{2\pi }}{\lambda }\) is the wave number of the ultrasound and a is a flaw size dimension. In order to use these algorithms effectively in the determination of flaw parameters, it has been found necessary to obtain measurements of the flaw at several viewing angles. At this time, there is no ultrasonic transducer available which permits this to be done efficiently and conveniently in the long and intermediate wavelength end of the spectrum. This region has been shown to be quite rich in flaw information and is appropriate to ultrasonic NDE in many practical applications (e.g., thick wall sections).

The development of a reliable eddy current inspection system to detect second layer cracks in sleeved engine disk bolt holes poses serious difficulties. This paper discusses some initial results obtained in two separate investigations that are aimed at advancing the state-of-the-art in eddy current detection of subsurface cracks. Both finite element design optimization results of a horseshoe shaped ferrite core probe, and the results of preliminary evaluation of the applicability of electric current perturbation (ECP) technique to the current problem are presented in this paper.

Using a two dimensional finite element code, the response of a U-core eddy current probe was computed for a subsurface flaw in a stainless steel medium. Next, using a three dimensional scattering model, the change in coil impedance was calculated for the same situation. From a comparison of these two results, it was concluded that the two dimensional finite element code overestimates the eddy current sensor response for the practical problem at hand by a factor of 10. This agreed well with the result obtained using an approximate technique described in this paper to estimate the true response from two dimensional calculations. Application of such desensitization factor should allow the two dimensional calculations to be effectively used in design studies.

The nondestructive determination of the compressive component of stress in various kinds of interference fasteners is a problem of considerable interest. As noted in the previous paper,1 none of the usual techniques employed for stress measurement is appropriate for this case. This paper reports results obtained upon Nitinol couplers utilizing ultrasonic reflectivity and interpretive concepts described in that paper as advanced by Haines.2 The reader is referred to a paper in the Proceedings of the 14th Symposium, Southwest Research Institute, for full details of the present work.3