Richard W. Martin

Quantitative imaging of Rayleigh wave velocity with a scanning acoustic microscope

An acoustic microscope operating with impulse excitation has been used to perform measurements of the Rayleigh wave velocity by measuring the time difference between the direct reflected signal and the Rayleigh wave signal. The accuracy and precision of the methodology have been examined by performing measurements at a single location on an elastically isotropic sample of E6 glass. The accuracy of the Rayleigh wave velocity measurement has been determined to be better than 0.5%. The measured Rayleigh wave velocity of (3035/spl plusmn/5) m/s differs by 0.3% from measurements reported in the literature for a similar sample, using two different techniques. The methodology has been extended to acquire the Rayleigh wave velocity while raster scanning the sample to develop a quantitative velocity image. The background noise in the Rayleigh wave velocity image has been investigated by mapping the velocity on elastically isotropic E6 glass. Possible reasons for background noise in the images is discussed. The methodology has been extended to acquire quantitative Rayleigh wave velocity images on Ti-6Al-4V. The contrast in the images is attributed to the variation of the Rayleigh wave velocity in individual grains or regions. Applicability of the technique to investigate crystallographic texture in materials is discussed.

Signal Processing of Leaky Lamb Wave Data for Defect Imaging in Composite Laminates

Inspection of composite laminates with Leaky Lamb waves (LLW) has been shown to hold promise of improved reliability and increased sensitivity to important defects [1]. Conventional scanning with the LLW has the possible disadvantage that the method is sensitive not only to internal structure, but also to small variations in plate thickness, which are indistinguishable from elastic property changes. To circumvent this potentially irrelevant sensitivity, a technique has been developed [2] whereby such variations can be selectively ignored, while retaining sensitivity to important defects or material property variations. The method consists of applying frequency modulation to the usual tone burst RF signal and exploiting detailed knowledge of the Lamb wave spectrum of composites [3] to discriminate between significant defects or property changes and small thickness variations in the plate. The current work extends and expands this analog signal processing scheme by performing the analysis on digitized data, permitting a much more general and flexible approach which will be described.

Local surface skimming longitudinal wave velocity and residual stress mapping

Local variation in surface skimming longitudinal wave (SSLW) velocity has been measured using a scanning acoustic microscope. A very narrow width electrical impulse has been used to excite the transducer of the acoustic lens. This permits the separation of the SSLW signal from the direct reflected signal in the time domain. A simple method of measuring the time delay between the directly reflected signal and the SSLW signal at two defocuses has been utilized for the local measurement of SSLW velocity. The variation in the SSLW velocity measured over an area of the sample is scaled and presented as an image. The method has been implemented to image the variation of the SSLW velocity around a crack tip in a sample of Ti-6Al-4V. Since the SSLW velocity is known to change linearly with the stress, the SSLW velocity image is considered as a representation of the image of stress around the crack tip. Local stress variation in the same region of the crack tip is directly measured using x-ray diffraction. The SSLW velocity image is compared with the x-ray diffraction stress image. The contrast in the two images, spatial resolution, and the penetration depth into the sample of acoustic waves and x rays are discussed.

Applications of Digital Image Enhancement Techniques to the Ultrasonic NDE of Composite Materials

Results of the applications of local and global digital image enhancement techniques to ultrasonic C-scan images of damaged graphite/epoxy composites are presented. The original unenhanced images were generated by using focused ultrasonic transducers with center frequencies between 3.5 and 25 MHz. Small defects were often difficult to detect in the unenhanced images because the relatively small signal amplitude changes resulting from the defects were obscured by the larger signal amplitude changes caused by variations in: (1) surface roughness, (2) material attenuation, and (3) material morphology. Results given in this paper indicate that those enhancement techniques which emphasize the higher spatial frequencies at the expense of the lower spatial frequencies and those techniques which operate on local pixel regions can often remove enough of the undesirable variations to make small defects visible in the enhanced images.

Imaging of Impact Damage in Composite Materials

Conventional ultrasonic C-scan images are generated from information acquired within “gates” placed at specific temporal locations on low-pass filtered and rectified versions of A-scans. Placing the gates at temporal locations which correspond with interfaces allows the integrity of the interfaces to be examined. However, if the interfaces are closely spaced, as is the case for quasi-isotropic graphite/epoxy composites, the information from upper layers is blurred into the layers below because of the finite time duration of the ultrasonic pulse. This creates a low signal-to-background-level ratio, which causes blurring at and below the first interface.

Development of a Scan System for Rayleigh Wave Velocity Mapping

A velocity map contains more quantitative information than an amplitude image because Rayleigh wave velocity is related to the elastic constants of the material. A high-precision scanning acoustic microscope system has been developed by the University of Dayton Research Institute with the capability to generate Rayleigh wave velocity map images of aerospace materials. The velocity map is presented in a C-scan format in order to visualize the velocity distribution in a material or around a defect. Rayleigh wave velocity is measured using a time-of-flight (TOF) technique. This system utilizes impulse excitation in order to separate the direct reflected signal and the Rayleigh wave signal in the time domain. Time differences between these signals at two defocus depths are used to calculate Rayleigh wave velocity in real-time and display a 2D x/y velocity map image during the scan. Velocity measurement accuracy is demonstrated to be better than 1%. Software techniques that were developed to improve time measurement accuracy will be quantitatively compared with the velocity in standard materials. Both measurement accuracy and standard deviation of experimental data are used as the basis of comparison for each investigated technique. Techniques discussed include improved peak detection, signal averaging requirements, digitization rate, and software gate placement. C- scan images of Rayleigh wave velocity maps of several materials will be presented. [1][2]

Rayleigh Wave Velocity Mapping Using Scanning Acoustic Microscope

In a Scanning Acoustic Microscope (SAM) amplitude of focused acoustic beam reflected by a sample is utilized to produce acoustic images and to measure local elastic property for effective nondestructive characterization of materials. The most important acoustic rays involved in both imaging and quantitative measurements in an acoustic lens are shown in Fig.1. The extra contribution to the reflected signal from the Rayleigh waves generated at the interface between water and the sample surface enhances the contrast in acoustic images. Amplitude acoustic images produced at a defocus are very effective in revealing the microstructure structure, surface and near surface defects, flaws, micro cracks etc. On the other hand an interference between the direct reflected ray (PO) and the Rayleigh ray (AB-BC-DE-EF) [see Fig.1] produces a V(z) curve which displays periodic minima as the distance between the lens and the sample is varied. The periodicity in the V(z) curve is directly related to the Rayleigh wave velocity. This makes an acoustic microscope a quantitative tool for measurement of local elastic property. Several methodologies have been developed to analyze the V(z) curve to obtain high accuracy in the measurement of Rayleigh wave velocity. A computationally intensive procedure with additional experimental data on a sample that doesn’pt support Rayleigh waves has been velocity with an accuracy of 1 part in 104 m/s. Although this tedious and time consuming procedure is very useful for high accuracy single location measurements, time necessary to produce an image of the variation Rayleigh wave velocity over an area becomes forbiddingly too large.

INVESTIGATION OF LOCAL ELASTIC PROPERTIES IN FRICTION STIR WELDED TI‐6AL‐4V USING SCANNING ACOUSTIC MICROSCOPY

Local changes in the microstructure and ultrasonic wave velocity variation across a friction weld in Ti‐6A1‐4V are investigated using scanning acoustic microscopy. Surface and bulk acoustic wave velocity and amplitude measurements performed across the weld are presented. The changes in the characteristics of the surface waves are related to the near surface microstructure in different parts of the weld. The bulk velocity and amplitude changes thru the thickness show bright and dark bands particularly in the nugget region. Possible reasons for formation of such bands are discussed. Application of acoustic microscopy to detect localized process induced defects in friction stir welds is discussed.

Challenges in detecting damage in the presence of microstructural inhomogeneties in a friction stir welded aluminum alloy for reusable cryotanks

Continuous real time structural health monitoring will be a requirement for future space launch missions. Reusable metallic cryotanks manufactured using Friction Stir Welding (FSW) technology for multiple missions, demands weld and microstructural integrity. The FS weld contains multiple interfaces and a variety of microstructures. To develop NDE-based health monitoring capability which detects damage and monitors the progression of damage, in the presence of these microstructural inhomogeneities, is a challenging task. Most structural health monitoring techniques are based on acoustic wave propagation. To design and develop efficient and optimized health monitoring capability based on acoustics, it is necessary to incorporate local elastic property variations that arise due to differences in the weld microstructure. These local elastic property changes across FSW regions have been measured using a focused acoustic beam. Measurements across the weld line show variations with a maximum change of 1% in the sound velocities. Macroscopic measurements of velocity of surface acoustic waves propagating across and also parallel the weld line in a large plate show significant variation. Experimental results of local and macroscopic sound velocity measurements from the changing microstructure along with their impact on the design of structural health monitoring system are discussed.

Acoustic Interferometer for Localized Rayleigh Wave Velocity Measurements

Two instrumentation systems for measurement of Rayleigh surface wave (RSW) velocity are described. The first system consists of a more conventional methodology using matched RF amplifiers and phase detector/mixer circuits. In the second system, a lock‐in amplifier, operating at high frequency, replaces the matched RF amplifiers and phase detector/mixer circuit, therefore simplifying the instrumentation. Both systems have been used to measure relative Rayleigh wave velocity using a cylindrically focused acoustic transducer consisting of three elements. A high‐precision relative velocity measurement of Rayleigh surface waves is performed by exciting the central element and one of the outer elements with a tone burst signal and measuring the phase difference between the two received signals.