No description available.

Abstract Tungsten carbide-cobalt powder is pressed before sintering into a compacted form using punches and a die cavity. After the powder has been pressed to a specific shape, it is sintered and shrinks a certain amount to a final size. To accommodate this shrinkage, the pressing tools are designed to a certain “shrinkage percentage” and thus the pressed component or compact is larger than the sintered component by that percentage amount. During the pressing process, there is a large amount of friction between the powder being compacted and the die cavity wall. To counter pressing friction, a lubricant is pre-mixed with the tungsten carbide powder. In the past at Powder Industries, the powder was mixed with wax and all of the tools were designed to a 20% shrinkage. In recent times, the wax in the powder has been replaced by PEG (polyethylene glycol) by most manufacturers as this increases the quality of the final product and is easier to remove in the furnaces. As a result of the new PEG lubricant, the tool wear rate at Powder Industries increased and because a higher pressure had been necessary to achieve powder pressing to the same shape and form, often the pressed components exhibited cracks or were not pressed ideally. On account of the problems introduced by PEG, correct tool design for the shrinkage was obtained by a ‘trial & error’ process. This project has been motivated by the need of establishing pressing and/or design ‘rules’ that would do away with trial and error when designing compaction tooling. The project has consisted of investigating the physical properties of 23 grades of WC-Co powder (with or without TiC and TaC) and of performing a series of pressing tests for each grade. A relationship between the apparent density of a powder and the ideal green density of the green compact pressed from the same powder has been found. Using this relationship, an equation has been derived between ideal shrinkage, powder apparent density, component sintered density and powder volatile content. Since the last three parameters are known to the tool designer, this equation can be used to calculate the ideal shrinkage when designing new compaction tooling. This method of calculating shrinkage is now in general use at Powder Industries and many successful sets of compaction tooling have already been manufactured

In most industrialized countries, significant future activity in the construction sector will be related to repair and rehabilitation of aging infrastructure. This will require use of durable and high performance repair materials. Among various mechanisms cited for lack of durability in repairs, early-age shrinkage cracking in overlay materials is of utmost importance. Fiber-reinforcement can be used to alleviate some of the issues related to plastic shrinkage. However, quantifying the performance of cement-based composites under restrained shrinkage conditions remains an issue. Various test techniques are available to measure free and restrained shrinkage, but do not simulate the real constraint imposed by the substrate on the overlay. In this dissertation, an innovative test method called the bonded overlay technique is described. An overlay of fiber-reinforced material to be tested is cast directly on a substrate, and the entire assembly is subjected to controlled drying. Cracking in the overlay is then monitored and characterized. During the development of this test method, instrumentation was included to enable measurement of the crack propagation rate using image analysis, evaporation rate, heat development, and strain using embedded sensors. Using the above technique, the effect of mix proportion including variables such as water-cement ratio (w/c), sand-cement ratio (s/c), and coarse aggregate content were studied. An increase in w/c from 0.35 to 0.6 significantly increased the total cracking. Addition of coarse aggregates reduced shrinkage cracking, however, for the range of s/c investigated, no definite conclusions could be drawn. Mixes with 0-20% fly ash and a 7 lit/m3 dosage of shrinkage reducing admixtures indicated no significant reduction in cracking. The influence of fiber geometry on cracking in overlays was also investigated. Fiber types included different sizes of polypropylene and cellulose fibers and one type of glass fiber (volume fraction ranging between 0-0.4%). Glass fibers at a small dosage of 0.1% were the most efficient fiber and completely eliminated cracking. Of the two field projects considered: one was a plaza deck at the UBC Aquatic Center, where cellulose fibers were used, and the second at the UBC ChemBioE building, where polypropylene fibers were used in slabs-on-grade. Both overlays were instrumented with strain sensors, data from which were monitored over the Internet. Results clearly indicated that fibers reduced the strain development in fiber-reinforced overlays when compared to un-reinforced overlays. An energy-based fracture model was proposed to predict maximum crack widths and in a second study, an equation was proposed to correlate early-age shrinkage and flexural toughness of cellulose fibers. In both models, a reasonable correlation with the test data was observed. In addition, factorial design method was used and a mathematical model was proposed to correlate different variables such as w/c, s/c, and fiber dosage.

In most industrialized countries, significant future activity in the construction sector will be related to repair and rehabilitation of aging infrastructure. This will require use of durable and high performance repair materials. Among various mechanisms cited for lack of durability in repairs, early-age shrinkage cracking in overlay materials is of utmost importance. Fiber-reinforcement can be used to alleviate some of the issues related to plastic shrinkage. However, quantifying the performance of cement-based composites under restrained shrinkage conditions remains an issue. Various test techniques are available to measure free and restrained shrinkage, but do not simulate the real constraint imposed by the substrate on the overlay. In this dissertation, an innovative test method called the bonded overlay technique is described. An overlay of fiber-reinforced material to be tested is cast directly on a substrate, and the entire assembly is subjected to controlled drying. Cracking in the overlay is then monitored and characterized. During the development of this test method, instrumentation was included to enable measurement of the crack propagation rate using image analysis, evaporation rate, heat development, and strain using embedded sensors. Using the above technique, the effect of mix proportion including variables such as water-cement ratio (w/c), sand-cement ratio (s/c), and coarse aggregate content were studied. An increase in w/c from 0.35 to 0.6 significantly increased the total cracking. Addition of coarse aggregates reduced shrinkage cracking, however, for the range of s/c investigated, no definite conclusions could be drawn. Mixes with 0-20% fly ash and a 7 lit/m3 dosage of shrinkage reducing admixtures indicated no significant reduction in cracking. The influence of fiber geometry on cracking in overlays was also investigated. Fiber types included different sizes of polypropylene and cellulose fibers and one type of glass fiber (volume fraction ranging between 0-0.4%). Glass fibers at a small dosage of 0.1% were the most efficient fiber and completely eliminated cracking. Of the two field projects considered: one was a plaza deck at the UBC Aquatic Center, where cellulose fibers were used, and the second at the UBC ChemBioE building, where polypropylene fibers were used in slabs-on-grade. Both overlays were instrumented with strain sensors, data from which were monitored over the Internet. Results clearly indicated that fibers reduced the strain development in fiber-reinforced overlays when compared to un-reinforced overlays. An energy-based fracture model was proposed to predict maximum crack widths and in a second study, an equation was proposed to correlate early-age shrinkage and flexural toughness of cellulose fibers. In both models, a reasonable correlation with the test data was observed. In addition, factorial design method was used and a mathematical model was proposed to correlate different variables such as w/c, s/c, and fiber dosage.

In most industrialized countries, significant future activity in the construction sector will be related to repair and rehabilitation of aging infrastructure. This will require use of durable and high performance repair materials. Among various mechanisms cited for lack of durability in repairs, early-age shrinkage cracking in overlay materials is of utmost importance. Fiber-reinforcement can be used to alleviate some of the issues related to plastic shrinkage. However, quantifying the performance of cement-based composites under restrained shrinkage conditions remains an issue. Various test techniques are available to measure free and restrained shrinkage, but do not simulate the real constraint imposed by the substrate on the overlay. In this dissertation, an innovative test method called the bonded overlay technique is described. An overlay of fiber-reinforced material to be tested is cast directly on a substrate, and the entire assembly is subjected to controlled drying. Cracking in the overlay is then monitored and characterized. During the development of this test method, instrumentation was included to enable measurement of the crack propagation rate using image analysis, evaporation rate, heat development, and strain using embedded sensors. Using the above technique, the effect of mix proportion including variables such as water-cement ratio (w/c), sand-cement ratio (s/c), and coarse aggregate content were studied. An increase in w/c from 0.35 to 0.6 significantly increased the total cracking. Addition of coarse aggregates reduced shrinkage cracking, however, for the range of s/c investigated, no definite conclusions could be drawn. Mixes with 0-20% fly ash and a 7 lit/m3 dosage of shrinkage reducing admixtures indicated no significant reduction in cracking. The influence of fiber geometry on cracking in overlays was also investigated. Fiber types included different sizes of polypropylene and cellulose fibers and one type of glass fiber (volume fraction ranging between 0-0.4%). Glass fibers at a small dosage of 0.1% were the most efficient fiber and completely eliminated cracking. Of the two field projects considered: one was a plaza deck at the UBC Aquatic Center, where cellulose fibers were used, and the second at the UBC ChemBioE building, where polypropylene fibers were used in slabs-on-grade. Both overlays were instrumented with strain sensors, data from which were monitored over the Internet. Results clearly indicated that fibers reduced the strain development in fiber-reinforced overlays when compared to un-reinforced overlays. An energy-based fracture model was proposed to predict maximum crack widths and in a second study, an equation was proposed to correlate early-age shrinkage and flexural toughness of cellulose fibers. In both models, a reasonable correlation with the test data was observed. In addition, factorial design method was used and a mathematical model was proposed to correlate different variables such as w/c, s/c, and fiber dosage. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

In addition to immediate elastic deformations, concrete undergoes time-dependent deformations that must be considered in design. Creep is defined as the time-dependent deformation resulting from a sustained stress. Shrinkage deformation is the time-dependent strain that occurs in the absence of an applied load. The total strain of a concrete specimen is the sum of elastic, creep, and shrinkage strains. Several test beams for the Pinner's Point Bridge have been produced by Bayshore Concrete Products Corp., in Cape Charles, VA. These beams feature high strength concrete mix designs with specified 28-day compressive strengths of 55.2 MPa (8,000 psi) and 69.0 MPa (10,000 psi). These test beams were equipped with thermocouples to track interior concrete temperatures, and vibrating wire gages placed at the center of prestressing to record changes in strain. Laboratory creep and shrinkage testing was conducted on specimens prepared with identical materials and similar mixture proportions to those used at Bayshore. The temperature profile from the test beams during steam curing was used to produce match-cured specimens for laboratory testing. Two match cure batches were produced, along with two standard cure batches. Creep specimens from each batch were placed in the creep room and loaded to 30 percent of their after-cure compressive strength. The creep room had a temperature of 23.0 ± 1.7 °C (73.4 ± 3 ºF) and relative humidity of 50 ± 4 %. Companion shrinkage specimens were also placed in the creep room. Measurements were taken on the creep and shrinkage specimens using a Whittemore gage. Four cylinders were also equipped with embedded vibrating wire gages (VWGs) so that the interior and exterior strains could be compared. The Whittemore and VWG elastic and creep strains were similar, while the VWGs recorded significantly less shrinkage. The measured creep and shrinkage strains were compared to seven different models to determine which model was the most accurate. The models considered were ACI 209, ACI 209 modified by Huo, CEB Model Code 90, AASHTO-LRFD, Gardner GL2000, Tadros, and Bazant B3. The ACI 209 modified by Huo was most accurate in predicting time-dependent strains. / Master of Science

Digitized by Kansas State University Libraries

Regression tasks belong to the set of core problems faced in statistics and machine learning and promising approaches can often be generalized to also deal with classification, interpolation or denoising problems. Whereas the most widely used classical statistical techniques place severe a priori constraints on the type of function that can be approximated (e.g. only lines, in the case of linear regression), the successes of sparse kernel learners, such as the SVM (support vector machine) demonstrate that good results may be obtained in a quite general framework by enforcing sparsity. Similarly, even very simple sparsity-based denoising techniques, such as classical wavelet shrinkage, can produce surprisingly good results on a wide variety of different signals, because, unlike noise, most signals of practical interest share vital characteristics (such as smoothness, or the ability to be well approximated by piece-wise linear polynomials of a low order) that allow a sparse representation in wavelet space. On the other hand results obtained from SVMs (and classical wavelet-shrinkage) suffer from a certain lack of interpretability, since one cannot straightforwardly attach probabilities to them. By contrast regression, and even more importantly classification, in a Bayesian context always entails a probabilistic measure of confidence in the results, which, provided the model assumptions are reasonably accurate, forms a basis for principled decision-making. The relevance vector machine (RVM) combines these strengths by explicitly encoding the criterion of model sparsity as a (Bayesian) prior over the model weights and offers a single, unified paradigm to efficiently deal with regression as well as classification tasks. However the lack of an explicit prior structure over the weight variances means that the degree of sparsity is to a large extent controlled by the choice of kernel (and kernel parameters). This can lead to severe overfitting or oversmoothing -- possibly even both at the same time (e.g. for the multiscale Doppler data). This thesis details an efficient scheme to control sparsity in Bayesian regression by incorporating a flexible noise-dependent smoothness prior into the RVM. The resultant smooth RVM (sRVM) encompasses the original RVM as a special case, but empirical results with a variety of popular data sets show that it can surpass RVM performance in terms of goodness of fit and achieved sparsity as well as computational performance in many cases. As the smoothness prior effectively makes it possible to use (highly efficient) wavelet kernels in an RVM setting this work also unveils a strong connection between Bayesian wavelet shrinkage and RVM regression and effectively further extends the applicability of the RVM to denoising tasks for up to millions of datapoints. We further discuss its applicability to classification tasks.

O presente estudo, in vitro, teve como objetivo avaliar a influência da variação da densidade de potência (300, 600, 1000 e 1550 mW/cm2) na contração de polimerização e na pigmentação de duas resinas compostas, Filtek Z250 e Filtek Z350XT (3M/ESPE). Para avaliação das forças geradas durante a contração de polimerização foram utilizadas duas bases de aço, em forma retangular (6 X 2 mm), posicionadas paralelamente na máquina de ensaios (Emic - DL 500), uma conectada à célula de carga de 100 Kg e, a outra, ao braço fixo. As resinas compostas foram introduzidas entre as bases metálicas, com altura de 1mm, volume de 12mm3 e fator C de 1,5. Para fotoativação foi utilizado o aparelho LED Blue Star 3 (Microdont Micro Usinagem de Precisão LTDA), determinando-se o tempo de 20 s para as diferentes densidades de potência. As bases ficaram fixas durante o ensaio (120 s) e as forças geradas durante a polimerização foram registradas numericamente e por meio de gráficos [força (N) x tempo (s)], num total de dez ensaios para cada grupo. Para os testes de pigmentação foram confeccionados, em uma matriz pré-fabricada, cinco espécimes circulares (6 mm X 2 mm) para cada grupo. As avaliações das alterações de cor (_E) foram realizadas com o Espectrofotômetro Easyshade com o sistema CIE-Lab, em diferentes períodos (inicial, após 6 e 12hs, 1, 2, 3, 4 e 5 dias de imersão em café). Os resultados obtidos foram submetidos à análise estatística ANOVA a dois critérios e teste Tukey para comparações individuais (p<0,05). Os resultados evidenciaram que a variação da densidade de potência influenciou diretamente na indução da contração de polimerização, com geração de forças significantemente maiores à medida que se aumentou os valores de irradiância, independente da resina composta utilizada. De forma semelhante, as alterações de cor apresentaram uma relação com a variação da densidade de potência, porém de forma indireta, evidenciando-se que o aumento progressivo da irradiância inferiu em menores valores de _E, assim como uma pigmentação mais acentuada para a resina composta nanoparticulada em relação à resina micro-híbrida. Considerando a observação de efeitos distintos com emprego de diferentes densidades de potência, dentro de um mesmo tempo de fotoativação das resinas compostas analisadas, tanto em relação à indução da contração de polimerização quanto de alteração de cor, deve ser concebido que o fator densidade de potência não pode ser estabelecido de forma independente aos demais parâmetros técnicos relacionados à dose de irradiância, à sequência de energia necessária e ao espectro de luz requerido para efetivação da polimerização das diferentes formulações de compósitos disponíveis. / The aim of this in vitro study was to evaluate the influence of the power density (300, 600, 1000 and 1550 mW/cm2) on polymerization shrinkage and pigmentation of two composite resins, Filtek Z250 and Filtek Z350XT (3M/ESPE). The experimental groups were set based on different combinations of the influencing factors. For the polymerization shrinkage test two steel bases with a parallel disposition were used, the movable one was connected to a 100 kg load cell and the other one was fixed using a universal testing machine (EMIC - DL 500). The resin composite was inserted between the bases and light-cured for 20 s with a LED unit (Blue Star 3, Microdont) The bases stayed fixed during the test (120 s) and the polymerization stresses generated were recorded in graphs [curves force (N) x time (s)]. A total of ten tests for each group were made. For the pigmentation tests five circular specimens (6 mm X 2 mm) for each group were made using a prefabricated mold. Using the color-measuring device Easyshade with the CIE-Lab system, evaluations were made at different times (baseline, after 6 hours, 12 hours, 1, 2, 3, 4 and 5 days of coffee immersion). Statistical analysis was performed with two-way ANOVA and Tukey test for individual comparisons (p <0.05). The results showed that the power density variation directly influenced the induction of polymerization shrinkage, with significantly greater generation force as it increased the values of irradiance, independent of the resin used. Similarly, color changes showed a relation with the power density variation, however demonstrating indirectly that the gradual increase in irradiance inferred at lower values of _E, as pigmentation was more pronounced for the composite with nanoparticles compared to a micro-hybrid resin. Considering the observation of various effects when using different power densities within the same light-curing time of the composite resins analyzed, in relation to the induction of polymerization shrinkage and color change, should be understood that the power density factor can not be established independently of other technical parameters related to the irradiation dosage, the sequence of necessary energy and light spectrum required to effecting the polymerization of composites with different available formulations.

Digitized by Kansas Correctional Industries