Abstract Accompany a rapid growth in the semiconductor industry in the past few year, most components gradually used the small dimension as its basic structures. Due to the reduction of component size will induces highly concentrated on circuit and dimension, it also incurs a lots problem, such as electromagnetic interference, high temperature and thermal stress, which will decrease the product reliability. The most common damage in the semiconductor product is thermal fatigue, which is caused by thermal stress concentrated under repeatedly temperature variation loading. Usually, the thermal cycle loading is applied to induce the fatigue destruction and predict the product reliability, but this method spends one cycle for 80min which is time-consumption. Therefore, in this thesis, the finite element method package is used to simulate and evaluate the plastic variation of solder bump and the relation between different temperatures loading and equivalent plastic strain under different temperature range test. Through the Coffin-Manson equation, the equivalent plastic strain can be used to predict the fatigue live, which can be precisely accelerating the fatigue test.

Flexural mechanism design is one of the latest field for mechanism researches in these recent years. In this paper, it will address one method of normalized dimensional parameters. According to the traditionally change of the appearance of flexural hinge, the different parameters will be designed through to the differences of several combination. On the other hand, we use the Finite Element Method to analyze the kinematics behavior of different hinges. Finally, we come up with a series of complete data that provide designers a reference for designing.

In order to understand the dynamic characteristics of an ultrasonic motor stator, we proposed a modified two- dimensional axisymmetric finite element model to analyze the three-dimensional vibrational problem of piezoelectric annular and circular plates. In this work, displacement fields are properly assumed and the electric effect is included. Following the finite element method, analyses of axisymmetric and nonaxisymmetric vibration of circular and annular plate, and also the stator of ultrasonic motors can be conducted in a convenient way. Natural frequency, location of contact point and elliptic locus of the stator are then calculated. Effects by different geometry design and selected circumferential wave number are discussed. Comparisons of some typical examples are made between the present work and those available in the literature.

The in-mold electromagnetic stirrer is a kind of device which is utilized to stir the molten steel in the steel factory. This thesis provides a detailed three-dimensional electromagnetic analysis of an in-mold electromagnetic stirrer driven by the moving magnetic field produced from stator winding currents. A commercial finite element analysis software will be utilized to calculate the flux density, eddy current, and electromagnetic force from static and dynamic analyses, and the above three physical phenomena are also discussed to obtain the 3-D electromagnetic characteristics. In order to improve the operational properties of the in-mold electromagnetic stirrer, the various position of the stator is modified to observe the distribution of the electromagnetic force. Besides, the magnitude and frequency of the input currents are also adjusted to predict the probable performances during on-site operation.

In the present study, effect of satisfying stress boundary conditions, in addition to displacement boundary conditions, in the axisymmetric vibration analysis of circular and annular plates is investigated. A new axisymmetric finite element, which is based on a combination of the conventional displacement-type variational principle and the Reissner¡¦s principle, is proposed. With this formulation, stresses, like displacements, are primary variables, and both displacement and stress boundary conditions can be easily and exactly imposed. Axisymmetric vibration frequencies of some typical circular and annular plates are then obtained with the present approach and are compared with those by the displacement-type axisymmetric finite element. It is found that the conventional finite element, though not satisfying stress boundary conditions, can still obtain sufficiently accurate vibration frequencies of circular and annular plates.

The current packaging trend toward to smaller and thinner has pushed the manufacturing technology to the limits. During the assembly processes of IC packages, delamination at interfaces and mechanical breakage of components are common mode of failure. The induced thermal stress within the package is one of the major contributions to these failures. According to the disparity of the coefficient of thermal expansion (CTE) between different components, internal thermal stress and warpage will be induced when the package undergoes temperature excursion. In this paper, the Die attach epoxy curing and encapsulation curing process induced warpage and thermal stress were studied by finite element software Marc & Mentat. As comparison, two kinds of Molding materials of the package considered, and the result will compared with the experimental data. Finally, studied the effect of the material thickness and the impact significance of each design variable on the design objective will also be discussed.

Abstract Using a finite element package software DEFORMTM, this study simulated plastic deformation of the sheet at the roll-gap during sheet rolling processes. Rigid-plastic model of material is adopted. The rolls are assumed to be perfectly rigid and the sheets are isotropic. This study consists of three parts¡G(1) analysis of asymmetrical sheet rolling ; the effects of rolls speed ratios, roll radii ratios, friction factor ratios between sheet and rolls surface, upon the curvature of the rolled product and rolling force were systematically discussed. The larger the roll radius and the roll speed ratios are, the smaller the rolling force is and the larger the curvature of the sheet analytically. (2) Closing behavior, the deformation mechanism and stress-strain distributions around internal voids inside the sheet during sheet rolling was discussed numerically. The influences of various rolling conditions, such as the thickness reduction, the dimension of the internal void, friction factor, cross-sectional area of the void, etc., on the dimension of the void at the exit were discussed. The critical thickness reductions, over which the void would close completely, were also investigated. It is known that the critical reduction decreases with increasing roll radius. That is because the rolling pressure at the roll gap increases with increasing roll radius, accordingly, it takes smaller reductions to make the void closed completely. The critical thickness reduction decreases with increasing roll radius and decreasing voids dimension-thickness ratios. (3) The possibility of voids occurred at the front and back of the inclusion was investigated. The effects of various rolling conditions, such as the thickness reduction, the diameter of the inclusions, the roll diameter, friction factor, etc., on the generation and development of voids anticipated to occur in the front and rear of the inclusions were discussed. The critical thickness reductions under which void generation can be avoided were also examined. On the other hand, the effect of friction factor on between roll and sheet upon the void lengths in the front and rear of the inclusion is greater than that of friction factor on between inclusions and sheet. It is known that it decreases with increasing inclusions dimension, whereas the critical reduction decreases with decreasing roll radius. The analytical results showed a steady trend for 0.1mm< voids and inclusions dimension < 2.5mm. To verify the validity of the models, a series of experiments on the sheet rolling using aluminum (A1050, A6061) sheet as specimen were carried out. The experiments on sheet rolling with an internal void inside the sheet were conducted by micro-drill. The inclusions used ¡§SUS HSS¡¨ steel. The void length in the front of the inclusion is larger than that in the rear of the inclusion for both simulated and experimental results. The simulated results were compared with experimental results, and good agreement is found. Therefore, this numerical model using DEFORM software can offer useful knowledge for designing the pass-schedule in sheet rolling processes.

Classical alternating direction (AD) methods for parabolic equations, based on some standard implicit time stepping procedure such as Crank-Nicolson, can have errors associated with the AD perturbations that are much larger than the errors associated with the underlying time stepping procedure . We plan to show that minor modifications in the AD procedures can virtually eliminate the perturbation errors at an minor additional computational cost. A mixed finite element method is applied in the spactial variables. Similar to the finite difference and finite element methods in spactial variables, we plan to have the same accuracy in time. A convergence analysis can also be shown .

The focus of the research is to develop a consistent and repeatable method to evaluate the coefficient of thermal expansion (CTE) of carbon fibers at high temperatures. Accurate measurement of the CTE of carbon fibers is essential to understand and develop optimal processing procedures as well as computational simulations to predict properties and allowables for fiber-reinforced composites. The mismatch between the coefficient of thermal expansion of the fiber and the matrix has a profound impact on the development of residual stresses and the subsequent damage initiation and progression, potentially diminishing the performance of composite structures. In situ transmission electron microscopy (TEM) is selected to perform the experimental work on account of the high resolution and the capability of evaluating both the longitudinal and transverse CTE. The orthotropy in the CTE is tested by rotating the fibers through 45° about their axis. The method is validated by testing standard tungsten filaments of known CTE. Additionally, the microstructure of the fibers is studied in a field emission scanning electron microscope as well as through selected area diffraction patterns in a TEM to observe presence of any potential orthotropy. The pitch based P55 fiber revealed a cylindrically orthotropic microstructure, but the PAN based IM7 and T1000 fibers did not reveal any orthotropy. Finite element models of hexagonally arranged IM7 fibers in a 977 epoxy matrix are developed using PATRAN and analyzed using the commercial FEA code ABAQUS 6.4. The fiber properties were considered temperature independent where as the matrix properties were varied linearly with temperature. The lamina properties evaluated from the finite element modeling are in agreement with the experimental results in literature within 10% in the temperature range of room temperature to the stress free temperature of the epoxy, however at cryogenic temperatures the difference is greater. The residual stresses developed during processing of the composite indicated a potential location for fiber matrix debonding to be in the matrix dominant regions.

¡@¡@The purpose of thesis is to study the responses of nanoindentation in fused silica. By experiments the mechanical properties of intrinsic fused silica were obtained. From the finite element simulation the response of material was estimated. Our main work is on simulation. This part includes the effects of different coefficient of friction, different indentation depth, tip rounding, and substrates of thin films. ¡@¡@First, the experimental load¡Vdisplacement curves were obtained through the nanoindentation sensing system. Then, a three-dimensional finite element was successfully modeled through the comparison of the load¡Vdisplacement curves of the experiment and the simulation. The yield stress and the strain-hardness trend of intrinsic fused silica were obtained. ¡@¡@For different coefficient of friction and different tip radii, no significant differences were found through the load¡Vdisplacement curves and von Mises stress distributions. For different indentation depths, varied trends were found through the load¡Vdisplacement curves and von Mises stress distributions. For substrate effect, no significant differences could be found through the normalized hardness. The intrinsic film hardness could be obtained for indentation depth less 20% of the total indentation depth.