Prasanna Karpur

Efficient use of Lamb modes for detecting defects in large plates

In this paper, Lamb wave propagation in large plates and its use in internal defect detection have been studied. The Lamb modes which are most efficient for detecting different types of internal defects are identified. Stress fields inside the plate for different Lamb modes are computed. From these stress plots one can conclude which Lamb mode should be efficient for detecting which type of material defect. Theoretical predictions are experimentally verified.

A Lamb wave scanning approach for the mapping of defects in [0/90] titanium matrix composites

In this paper a new scanning technique using leaky Lamb waves is presented. This technique is applied to detect internal defects in a multilayered fiber-reinforced composite plate specimen (SCS-6 fibers in Ti-6Al-4V matrix). Images generated by this new Lamb wave scanning technique (we will refer it as the L-scan technique) are compared with conventional C-scan images. This comparison shows that the L-scan technique is more effective for detecting some internal defects such as missing fibers and fiber breakage type defects in a multilayered specimen than the conventional C-scan technique.

Ultrasonic reflectivity technique for the characterization of fiber‐matrix interface in metal matrix composites

An ultrasonic plane wave reflected by a cylindrical fiber embedded in a homogeneous isotropic matrix is modeled. The model calculates the ‘‘back‐reflection’’ coefficient by taking in to account the properties of the fiber and the matrix, the ultrasonic wavelength, the angle of incidence, and a coefficient called ‘‘shear stiffness coefficient’’ which characterizes the elastic behavior between the fiber and the matrix. Results obtained from the theoretical analysis for a model metal matrix composite system are shown. The theory developed in this paper and some of the results obtained are equally applicable in ceramic matrix fiber reinforced composites.

Split spectrum processing : a new filtering approach for improved signal-to-noise ratio enhancement of ultrasonic signals

Abstract Split spectrum processing is a frequency diversity technique used to enhance the signal-to-noise ratio of ultrasonic signals. Traditionally, many Gaussian bandpass filters are used to achieve frequency diversity. This paper introduces an alternative method of obtaining frequency diversity wherein raised-cosine bandpass filters with flat-tops are recommended for use. The reasoning behind the use of raised-cosine filters is outlined. Experimental results are provided to demonstrate the advantages of the raised-cosine filters over Gaussian filters. Finally, the limitation of the new filters is experimentally shown.

Measurement of the dynamic elastic moduli of porous titanium aluminide compacts

The dynamic elastic moduli of the porous alpha-two titanium aluminide compacts are measured using an ultrasonic technique. Both shear and longitudinal velocities are measured for compacts of different densities, making computation of all the four elastic constants, namely, the Young’s modulus, shear modulus, bulk modulus and Poisson’s ratio. The dependence of these on the relative density are correlated and compared with some earlier models, and some of the uncertainties in the earlier models are discussed.

A self-learning neural net for ultrasonic signal analysis

Abstract This paper deals with ultrasonic signal analysis using artificial neural nets. In particular, it investigates the classification of ultrasonic inspection data using a backpropagation (BP) neural network. The traditional BP learning algorithm requires a set of signals that have been classified a priori for training the net. To eliminate the uncertainties involved in selecting the training set, a self-learning algorithm is developed. A BP net is trained using the self-learning algorithm and it is used to classify ultrasonic inspection data. The classification is performed in both the time and the frequency domains. The classification results are compared with the results obtained by conventional C-scan methods, and it is found that 95% of all the signals with flaws are classified correctly by the unsupervised BP net. The self-learning algorithm developed for classification of ultrasonic data is also detailed in this paper. It utilizes the correlation information computed by the first hidden layer of the BP net for the generation of the initial training set consisting of one signal per each class. Additional signals in each class are selected by the net using the nearest neighbour approach. An estimate of the number of different classes present in the data is made using the damage profile of the sample being investigated by Kohonens learning vector quantization (LVQ) algorithm.

In situ observation of the single-fiber fragmentation process in metal-matrix composites by ultrasonic imaging

Abstract Single-fiber fragmentation tests with continuous silicon-carbide fibers in a Ti6Al4V alloy matrix have been conducted with in situ ultrasonic imaging to monitor the fragmentation process. Straining proceeded incrementally on a specially designed load frame with acoustic emission detection (AE) performed during each increment, and shear-wave back reflectivity (SBR) ultrasound images were acquired following each increment. Metallographic examination of the fragmented fiber was performed following the straining sequence by electropolishing and scanning electron microscopy. Good agreement was found between the fiber breaks imaged by ultrasound, the number of breaks detected by acoustic emissions, and the breaks observed by metallography.

Fracture strength and damage progression of the fiber/matrix interfaces in titanium-based MMCs with different interfacial layers

Abstract In this paper, a concerted utilization of finite element analysis and an ultrasonic characterization technique is described to assess the interfacial fracture strength and to monitor the progression of damage at the interfacial region in titanium-based metal-matrix composites. The finite element model developed here encompasses an interfacial element with a finite thickness to simulate the interfacial region of the coating or reaction products. The finite element model has been used in conjunction with the ultrasonic evaluation technique to assess the in situ interfacial fracture strength. The different responses of the ultrasonic amplitudes for Ti-6A1-4V/SCS-0 SiC and Ti-6A1-4V/ SCS-6 SiC interfaces have been explained in terms of the reflection of ultrasonic waves from the fiber/matrix interface. It is established that the non-monotonic stress dependence of the ultrasonic reflection amplitude for both the SCS-0 and SCS-6 interfaces is related to the debonding between the fiber and matrix. The results indicate that the SCS-0 interface has a much higher fracture strength than the SCS-6 interface although both these interfaces exhibit similar apparent debonding stresses.

Ultrasonic characterization of the fiber-matrix interphase/interface for mechanics of continuous fiber reinforced metal matrix and ceramic matrix composites

Abstract This paper presents a novel approach to evaluate the elastic properties and the behavior of the interphase region formed by a chemical reaction between the matrix and the fiber materials in metal matrix and ceramic matrix composites. Contrary to the traditional approach which does not allow any relative displacement at the interface without fracture, this paper considers elastic deformation of the interphase zone between the matrix and the fiber by replacing the zone by an “equivalent elastic interface”. The elastic behavior of the equivalent elastic interface describes the local elastic rigidity and deformation of the interphase zone and can be quantified by a mechanics parameter called “shear stiffness coefficient” which is proportional to the ratio of the shear modulus to the local thickness of the interphase material. This paper also outlines an ultrasonic reflectivity modeling that can be used for the experimental measurement of the interfacial shear stiffness coefficient along the length of an embedded fiber. Further, an experimental method of measurement of the shear stiffness coefficient is presented and experimentally measured values are tabulated. The significance of the quantification of such a parameter is that the elastic property of the interface obtained can be used as a common basis among material scientists designing and developing the composite systems, and groups studying material behavior for life prediction. Also, the parameter can be used by production engineers to assure that the designed properties of the composite are being achieved, and by the end users to ensure that the designed and produced properties are being retained in use.

Adhesive Joint Evaluation Using Lamb Wave Modes with Appropriate Displacement, Stress, and Energy Distribution Profiles

One of the most elusive yet critical problem in adhesive joints characterization is that of ‘kissing bond’ wherein good contact exists among the adherend and the adhesive, however with no acceptable levels of adhesion. To date, the kissing bond is difficult to be detected reliably by any of the methods including conventional ultrasound and thermal waves. Kissing bond which is a manufacturing defect/anomaly will substantially compromise the load bearing capability of the adhesive joint by initiating adhesive failure (in contrast to cohesive failure wherein the failure occurs within the thickness of the adhesive layer instead of a failure at the interface). Attempts to develop methods of detection of kissing bonds have been unsuccessful to date.