The objective of this research is the pursuit of a better characterization method for the air voids - particularly air-entrained voids - in fresh cement-based materials through the use of ultrasonics. The use of air-entraining agents (AEA's) to incorporate a stable network of air voids into fresh cement paste is common practice in the concrete industry. These particular air voids significantly improve durability of hardened cement paste through mitigation of stresses associated with freezing and thawing cycles. It is understood that the performance of entrained air voids in cement paste is dependent on their size and distribution, or spacing factor. However, conventional methods for qualifying air content, such as the Pressure, Volume, and Gravimetric Methods, only measure total air volume and cannot assess size or spacing. In this investigation, using matched pairs of transducers, ultrasonic pulses were transmitted through fresh cement paste specimens (0.0\% up to 0.6\% AEA by weight of cement). The received signals were recorded every five minutes during the first six hours and then every fifteen minutes thereafter. Analysis shows strong distinctions between specimens with and those without the AEA. Further research is needed into the distinctions among specimens with the AEA. However, the data suggest correlations between Vicat setting times, heat of hydration, and autogenous strain and ultrasonic metrics such as pulse velocity and peak frequency of the signal. The findings of this research should be most appropriate as a foundation for an inversion process and improved air-entrainment detection methods.

In this thesis ultrasonic Lamb wave measurements are performed to detect material nonlinearity in aluminum sheets. When a Lamb wave propagates, higher harmonic wave fields are generated and under certain conditions the second harmonic is cumulative. When these conditions hold the Lamb waves are serviceable for material nonlinearity measurements. For generation, a wedge transducer combination is used. The detection of the Lamb wave are performed with either a laser interferometer or a second wedge transducer combination and the results are benchmarked. A short time Fourier transformation (STFT) is applied to the detected signal to extract the amplitudes of the first and second harmonics. A relative ratio of the first and second harmonics is deduced from nonlinear wave theory to assign the nonlinearity of the material. To verify the capability of the measurement setup and to show that cumulative second harmonics are generated, measurements for different propagation distances are performed. Further measurements on plasticly deformed specimens are carried out to examine the change of the material nonlinearity as a function of plasticity.

The objectives of this thesis are to develop and validate laser ultrasonic inspection methods for on-line testing of microelectronic devices. Electronic packaging technologies such as flip chips and BGAs utilize solder bumps as electrical and mechanical connections. Since they are located hidden from view between the device and the substrate, defects such as cracks, voids, misalignments, and missing bumps are difficult to detect using non-destructive methods. Laser ultrasonic inspection is capable of detecting such defects by utilizing a high power laser pulse to induce vibrations in a microelectronic device while measuring the out of plane displacement using an interferometer. Quality can then be assessed by comparing the vibration response of a known-good device to the response of the sample under inspection. The main limitation with the implementation of laser ultrasonic inspection in manufacturing applications is the requirement to establish a known-good reference device utilizing other non-destructive methods. My work will focus on developing a method to inspect flip chip devices without requiring a previously established reference. The method will automatically examine measurement data from a large sample set to identify those devices which are most similar. The selected devices can then be utilized to compose a hybrid reference signal which can be used for comparison and defect detection. Current trends in the electronic packaging industry continue to drive toward increased solder bump density, making it increasingly difficult to generate strong ultrasonic signals in these stiffer devices. To overcome this difficulty, I propose a new excitation method which places the source of ultrasound at the inspection location for each test point on the device surface. This ensures that the same power is available for each inspection location while also increasing the signal to noise ratio. The hardware implementation of this method reduces the system complexity and required automation, which can significantly reduce equipment cost and inspection time. The implementation of the proposed excitation method in conjunction with the use of a hybrid reference signal for defect detection will improve the utility of the laser ultrasonic inspection technique to on-line inspection applications where no other non-destructive methods are currently available.

The long-term goal of this research is to assist in the development of a fast, accurate, and low-cost nondestructive inspection prototype for solder joints in integrated circuits (IC). The goal of the work described in this thesis is to develop a fully automated system to maintain the signal strength of the vibrometer that would reduce the testing time while maintaining or improving the quality of the defect detection results. The ability to perform the inspections in an automated manner is very important in order to demonstrate the ability of the defect detection system to be used for online inspection without the need of an operator. The system was able to find the maximum signal strength (at a single point on the surface of a flip chip) nearly five times faster than Polytec's commercial system with a search time of approximately 2.1 sec. When integrated into the nondestructive inspection prototype, the system described in this work was found to approximately reduce the data acquisition time per test location by four times, with a minimum data acquisition time of 8.5 sec and an average time of 15.4 sec, while maintaining the same level of quality of results obtained by a skilled operator when manually maintaining the signal strength of the vibrometer. Hardware was developed that retrofitted a vibrometer's focusing head at the end of a fiber optic cable to a motorized linear stage. This stage controlled the standoff distance between the focusing head and the IC's surface with a fixed focal length, which allowed the spot size of the laser to be adjusted while searching for a desired signal strength. Numerous tests were conducted to determine the search parameters, which led to a search time of approximately 2.1 sec. This time was found to be dependent on the surface finish of the IC being inspected. It was also found that to achieve a desired signal intensity strength, not only does the standoff height of the focusing head, which determines the laser spot size, need to be controlled, but also the exact location on which the laser is reflecting off the IC.

Thesis (Ph.D.)--Tufts University, 2004. / Adviser: Masoud Sanayei. Submitted to the Dept. of Civil Engineering. Includes bibliographical references (leaves 300-305). Access restricted to members of the Tufts University community. Also available via the World Wide Web;

Thesis (M.S. in Civil Engineering)--University of Maryland, College Park, 1992. / "A scholarly paper submitted to Assistant Professor Ian Flood." Description based on title screen as viewed on April 16, 2009. Includes bibliographical references (p. 49-52). Also available in print.

Made available in DSpace on 2014-10-09T12:36:47Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:57:29Z (GMT). No. of bitstreams: 1 12897.pdf: 1698549 bytes, checksum: c18bc00ac97b58dffe2313af7761ee1d (MD5) / Dissertacao (Mestrado) / IEA/D / Escola Politecnica, Universidade de Sao Paulo - POLI/USP

Os ensaios não destrutivos servem como uma importante ferramenta para a análise de estruturas de concreto armado. A utilização de ensaios de velocidade de propagação do pulso ultra-sônico (VPU) permite realizar um acompanhamento das características do material ao longo de sua vida útil. Através da análise dos dados obtidos, pode-se averiguar a uniformidade do concreto, controlar a sua qualidade, acompanhar sua deterioração e, através de comparação com corpos de prova de referência e, até mesmo, estimar a resistência do mesmo. No entanto, as técnicas atuais para análise dos resultados coletados são, em grande parte, baseadas na sensibilidade dos profissionais que as aplicam. Para facilitar o controle e inspeção de estruturas de concreto armado é fundamental desenvolver estratégias para tornar esta análise mais simples e precisa. Este trabalho se baseia na hipótese de que a aplicação de Redes Neurais Artificiais (RNAs) pode gerar modelos de relacionamento úteis e acurados entre as características do concreto, sua compacidade e sua resistência à compressão. O intuito é determinar se com o uso de RNAs é possível estabelecer relações não-lineares que permitam estimar a resistência do concreto a partir do conhecimento de algumas propriedades básicas e da verificação da sua compacidade por meio de ensaios de VPU. Os resultados indicam que as RNAs podem ser usadas para gerar métodos numéricos robustos e flexíveis para estimativa da resistência à compressão a partir de dados de VPU. O estudo evidencia uma considerável melhora nos resultados de estimação da resistência quando se empregam modelos neurais, em comparação a modelos estatísticos tradicionais. Para os dados coletados, provenientes de diversas pesquisas, os modelos tradicionais geram estimativas com coeficientes de determinação que não ultrapassam um valor de R² de 0,40. Já as redes neurais conseguem ajustes com R² da ordem de 0,90. Além de contribuir para uma melhor análise de situações em que haja dúvidas sobre a resistência ou homogeneidade de elementos de concreto, o trabalho demonstra que modelos neurais são uma forma eficiente de ordenar e transferir conhecimento não estruturado. Constatou-se, ainda, que, dada sua capacidade de aprendizagem e de generalização do conhecimento adquirido, as RNAs se constituem em um meio rápido e preciso para modelagem de fenômenos complexos. / Nondestructive Testing (NDT) techniques are useful tools for analyzing reinforced concrete (RC) structures. The use of Ultrassonic Pulse Velocity (UPV) measurements enables the monitoring of changes in some critical characteristics of concrete over the service life of a structure. The interpretation of the data collected allows an assessment of concrete uniformity, and can be used to perform quality control, to monitor deterioration and even, by means of comparison against reference samples, to estimate compressive strength. Nonetheless, the current techniques for UPV data analysis are, on a large degree, based on the sensitivity of the professionals who apply these tests. For accurate diagnosis it is necessary to consider the various factors and conditions that can affect the results. To proper control and inspect RC facilities it is essential to develop appropriate strategies to make the task of data interpretation easier and more accurate. This work is based on the notion that using Artificial Neural Networks (ANNs) is a feasible way to generate workable estimation models correlating concrete characteristics, compacity and compressive strength. The goal is to determine if it is possible to establish models based on non-linear relationships that are capable of estimating with good accuracy the concrete strength based on previous knowledge of some basic material characteristics and UPV measurements. The study shows that this goal is achievable and indicates that neural models perform better than traditional statistical models. For the data collected in this work, provided by various researchers, traditional regression models cannot exceed R² = 0.40, while the use of ANNs allows the creation of models that can reach a determination coefficient R² = 0.90. The results make clear that, besides contributing to better the analysis of situations where there is doubts regarding concrete strength or uniformity, neural models are an efficient way to order and transfer unstructured knowledge. It was shown that, given the learning capacity and its ability to generalize acquired information into mathematical patterns, ANNs are a quick and adequate way to model complex phenomena.

In recent decades techniques related to the measurement of elastic waves have advanced significantly. It is now possible with relatively inexpensive equipment to record amplitudes and frequency bands that were unthinkable two decades ago. This has led to the development of theoretical topics which application was questionable until not long ago, to profit from new technological potential in obtaining more and better experimental information. In this context the study of the propagation of guided waves in solids is presented as a knowledge that allows to detect damage with efficiency and economy in a number of structures in which at least one dimension is much larger than the other two. This is the case for rails, tubulations and pressure vessels among others. In this work, guided waves propagation characteristics are studied, first in a prismatic bar and then within the geometry of a rail. In both cases, dispersion curves were computed according two different work methodologies, first using an axisymmetric model and then corroborated with a model of periodic conditions. Finally propagation of a Tone-Burst waves were simulated on the analyzed geometries, leading to the discussion of how the waves scatter along its propagation. The results obtained were the dispersion curves of both sections. / En las últimas décadas técnicas relacionadas con la medición de ondas elásticas han avanzado sensiblemente. Actualmente, con equipamientos relativamente económicos es posible registrar amplitudes y franjas de frecuencia que eran impensables dos décadas atrás. Este hecho ha motivado que tópicos teóricos que hasta hace un tiempo tenían una aplicación cuestionable tengan que ser desarrollados para poder aprovechar las nuevas potencialidades tecnológicas en la obtención de mayor y mejor información experimental. En este contexto, el estudio de la propagación de ondas guiadas en sólidos se presenta como un conocimiento que posibilita detectar daño con eficiencia y economía en una serie de estructuras en las cuales por lo menos una dimensión es mucho mayor que las otras dos. Es el caso de estructuras tubulares, rieles o recipientes sometidos a presión entre otras. En el presente trabajo se estudian las características de propagación de ondas guiadas primeramente una barra rectangular y posteriormente en la geometría de un riel. En ambos casos, fueron calculadas las curvas de dispersión obtenidas con por dos metodologías de trabajo por elementos finitos, la primer metodología fue emplear un cálculo aplicando un modelo axisimétrico, y luego corroborado con un modelo de condiciones periódicas y posteriormente fue simulada la propagación de una onda tipo Toneburst sobre las geometrías analizadas discutiendo cómo dicha onda se dispersa durante su propagación. Los resultados obtenidos fueron las curvas de dispersión de ambas secciones.

Os ensaios não destrutivos servem como uma importante ferramenta para a análise de estruturas de concreto armado. A utilização de ensaios de velocidade de propagação do pulso ultra-sônico (VPU) permite realizar um acompanhamento das características do material ao longo de sua vida útil. Através da análise dos dados obtidos, pode-se averiguar a uniformidade do concreto, controlar a sua qualidade, acompanhar sua deterioração e, através de comparação com corpos de prova de referência e, até mesmo, estimar a resistência do mesmo. No entanto, as técnicas atuais para análise dos resultados coletados são, em grande parte, baseadas na sensibilidade dos profissionais que as aplicam. Para facilitar o controle e inspeção de estruturas de concreto armado é fundamental desenvolver estratégias para tornar esta análise mais simples e precisa. Este trabalho se baseia na hipótese de que a aplicação de Redes Neurais Artificiais (RNAs) pode gerar modelos de relacionamento úteis e acurados entre as características do concreto, sua compacidade e sua resistência à compressão. O intuito é determinar se com o uso de RNAs é possível estabelecer relações não-lineares que permitam estimar a resistência do concreto a partir do conhecimento de algumas propriedades básicas e da verificação da sua compacidade por meio de ensaios de VPU. Os resultados indicam que as RNAs podem ser usadas para gerar métodos numéricos robustos e flexíveis para estimativa da resistência à compressão a partir de dados de VPU. O estudo evidencia uma considerável melhora nos resultados de estimação da resistência quando se empregam modelos neurais, em comparação a modelos estatísticos tradicionais. Para os dados coletados, provenientes de diversas pesquisas, os modelos tradicionais geram estimativas com coeficientes de determinação que não ultrapassam um valor de R² de 0,40. Já as redes neurais conseguem ajustes com R² da ordem de 0,90. Além de contribuir para uma melhor análise de situações em que haja dúvidas sobre a resistência ou homogeneidade de elementos de concreto, o trabalho demonstra que modelos neurais são uma forma eficiente de ordenar e transferir conhecimento não estruturado. Constatou-se, ainda, que, dada sua capacidade de aprendizagem e de generalização do conhecimento adquirido, as RNAs se constituem em um meio rápido e preciso para modelagem de fenômenos complexos. / Nondestructive Testing (NDT) techniques are useful tools for analyzing reinforced concrete (RC) structures. The use of Ultrassonic Pulse Velocity (UPV) measurements enables the monitoring of changes in some critical characteristics of concrete over the service life of a structure. The interpretation of the data collected allows an assessment of concrete uniformity, and can be used to perform quality control, to monitor deterioration and even, by means of comparison against reference samples, to estimate compressive strength. Nonetheless, the current techniques for UPV data analysis are, on a large degree, based on the sensitivity of the professionals who apply these tests. For accurate diagnosis it is necessary to consider the various factors and conditions that can affect the results. To proper control and inspect RC facilities it is essential to develop appropriate strategies to make the task of data interpretation easier and more accurate. This work is based on the notion that using Artificial Neural Networks (ANNs) is a feasible way to generate workable estimation models correlating concrete characteristics, compacity and compressive strength. The goal is to determine if it is possible to establish models based on non-linear relationships that are capable of estimating with good accuracy the concrete strength based on previous knowledge of some basic material characteristics and UPV measurements. The study shows that this goal is achievable and indicates that neural models perform better than traditional statistical models. For the data collected in this work, provided by various researchers, traditional regression models cannot exceed R² = 0.40, while the use of ANNs allows the creation of models that can reach a determination coefficient R² = 0.90. The results make clear that, besides contributing to better the analysis of situations where there is doubts regarding concrete strength or uniformity, neural models are an efficient way to order and transfer unstructured knowledge. It was shown that, given the learning capacity and its ability to generalize acquired information into mathematical patterns, ANNs are a quick and adequate way to model complex phenomena.