Thin films of metals and ceramics are commonly used as the structural materials for microelectromechanical systems (MEMS). These systems are used for a wide range of applications that include sensors, displays, and portable power generators. Accurate measurement of the mechanical properties of these thin films is essential for the robust design of high-performance and reliable MEMS. In this thesis, the method of nanoindentation is used to characterize the elastic properties of thin films of Hexoloy-SG silicon carbide. This material is of particular interest for MEMS operating in harsh, hightemperature environments. Nanoindentation was performed using a commercial Hysitron TriboIndenter® equipped with a diamond Berkovich nanoindenter tip. During each nanoindentation test, the indenter was forced into the specimen by a calibrated load, while the indentation depth was recorded continuously with nanometer resolution. The first part of this thesis describes a detailed protocol for nanoindentation testing using the TriboIndenter® and discusses the calibration of the instrument using a quartz standard. The identification of errors due to drift, vibrations, and surface inhomogeneities is described in detail. This protocol was then used to test a 2.1 µm thick film of Hexoloy-SG silicon carbide film deposited on a 500 µm thick single-crystal silicon substrate. The film was tested with loads varying from 1,000 µN to 13,000 µN, such that the maximum indentation was always less than 10% of the film thickness. The nanoindentation load-displacement curves from a total of 155 individual indents were analyzed using three different methods to extract the value of the Young's modulus of the Hexoloy film. The first method, developed by Oliver and Pharr, is widely used in nanoindentation analysis, but does not account explicitly for the effects of the underlying substrate. It is valid strictly for monolithic materials. Therefore, refinements of the Oliver-Pharr method by King a / Des structures formées de couches minces métalliques et céramiques sont couramment utilisées dans la conception de microsystèmes électromécaniques (MEMS). Ces derniers se retrouvent dans plusieurs domaines, tels que les capteurs, les vidéoprojecteurs et les systèmes de génération d'énergie portable. Pour concevoir des MEMS fiables, les propriétés mécaniques de ces couches minces doivent êtres connues précisément. Le but de cette thèse est d'utiliser la méthode de nanoindentation pour déterminer les propriétés mécaniques des couches minces de carbure de silicium Hexoloy-SG. Ce matériel à été développé pour des microsystèmes opérant dans des conditions thermiques et chimiques extrêmes. La nanoindentation a été réalisée par le système TriboIndenter® de Hysitron équipé d'une pointe Berkovich en diamant, de forme pyramidale à base triangulaire. Chaque indentation comprend un cycle charge/décharge durant lequel la pointe indentatrice est enfoncée et retirée du matériel par une force calibrée, tandis que la profondeur de l'indentation est surveillée continuellement au nanomètre près. La première partie de cette thèse décrit une procédure détaillée pour la nanoindentation avec le système TriboIndenter®, incluant la calibration de l'instrument utilisant une norme en quartz monolithique. Également, l'identification d'erreurs expérimentales reliées au système de nanoindentation dues à la dérive, aux vibrations et à la rugosité de l'échantillon et leurs mesures correctrices sont présentées. Par après, cette méthodologie a été utilisée pour tester les propriétés mécaniques d'une couche mince de Hexoloy-SG mesurant 2.1 µm d'épaisseur, déposée sur un substrat de silicium monocristallin comptant 500 µm d'épaisseur. La force exercée par l'indentateur sur l'échantillon varie de 1,000 µN à 11,000 µN, pour que la profondeur de l'indentation demeure en deçà de 10% de l'épaisseur totale de la co

Computer simulations play a significant role in the development and operation of space systems because of the difficulty in performing ground-based hardware testing and onorbit tests are impossible before the launch of these systems. Hence, accurate modeling and simulation of space robotic tasks involving contact is very crucial. This in turn implies that accurate model (contact) parameters, used as inputs to the software to represent the contact operation being simulated, are imperative. In this work, we addressed the contact parameter estimation problem for simple contacting geometries (one-point contact) and complex contacting geometries (multiple-point contact). Several frequency domain identification strategies were applied to one-point contact parameter estimation problem to estimate contact stiffness and damping. The performance of these frequency domain algorithms was evaluated and compared with time domain identification algorithms. The identification algorithm for multiple point contact scenario to estimate contact stiffness, damping and coefficient of friction was also investigated. We determined the cause for poor estimation results obtained with previous implementation of this identification algorithm and were able to improve the performance of the algorithm. Finally, a thorough evaluation of sensitivity of the algorithm to noise in measured data was conducted.

This thesis addresses calibration of joint torque sensors and identification of a gravity compensation (GComp) model. The two problems are related: calibrated torque information is prerequisite for the GComp model identification; the identified GComp model makes on-line automatic torque sensor calibration possible. / In the first part, we propose an autonomous joint torque sensor calibration method, which utilizes combinations of single-joint rotations and an arm's own gravity load. The method determines not only joint torque sensor gains and offsets, but also those of the joint angle sensors. When one joint of a manipulator is rotated, the gravity torque exerted on the joint varies sinusoidally with rotation angle. From the fitted parameters one may extract sensor gains and offsets. A custom designed calibration load is used to provide a reference torque for torque sensor gain calibration. A key feature of the proposed method is that nothing is assumed known about the arm's inertial parameters or of the location of the reference load in the grasp. Experiments are conducted on the Sarcos Dextrous Arm. / The second part of the thesis presents a new and simple procedure to identify link mass parameters used for gravity compensation and on-line automatic torque sensor calibration for a robot manipulator. The approach employs a single-joint rotation and a recursive procedure that proceeds distally to proximally. Composite mass moments are identified for each link, and are further divided into configuration dependent and independent terms. It is shown that the configuration-independent terms may actually combine contributions from several links. Simulation and experimental results on the Sarcos Dextrous Arm verify the approach.

Fluid passing through a cylindrical shell causes both lateral displacements and deformation of the cross-section. Studies have shown that this applies to both the cantilevered and clamped-clamped shells. The effect of boundary conditions was examined and it was concluded that a system with identical end conditions (clamped or pinned) is conservative; thus, at sufficiently high flow, it loses stability by divergence, followed by coupled-mode flutter at a slightly higher flow velocity. / This thesis focuses on the systems of clamped-pinned and pinned-clamped shells. Initially using two different theories, the standing wave analysis and the travelling wave analysis, small dampings (negative or positive) are found in a system of clamped-pinned (and pinned-clamped) ends before the flow reaches the critical divergence velocity. The same result is obtained based on an analysis of an elastically constrained shell. / Upon further examination, it is found that the small dampings observed in the first two cases are simply numerical artifacts: the magnitude of these dampings tends to zero as the accuracy of the computation is improved Hence, it is concluded that the clamped-pinned and pinned-clamped systems are conservative.

The thesis centers on time domain modelling of viscoelastic materials. Classical models are compared to models involving fractional derivatives, which are derivatives of an order between 0 and 1. Parameters for classical and fractional order models are found for two materials, polymethylmethacrylate and 3M ISD 112, an acrylic based material sold as a viscoelastic layer by 3M. In both cases, only Prony series with several parameters achieve a good representation over a large frequency range. In the case of 3M ISD 112, a fractional model with only two parameters gives a good representation over a frequency range of three decades, which is often sufficient. / An algorithm based on an approximated definition of the fractional derivative and a trapezoidal rule is described to solve constitutive equations with fractional derivatives. The algorithm is implemented in C and tested against a numerical Laplace inverse for the case of a material submitted to sinusoidal strains. The algorithm gives accurate results and does not require very small steps, which is usually the case for algorithms based on finite differences or Grunwald series. / The algorithm is adapted to the structure of a user subroutine of a commercial finite element package, Samcef, for a six component isotropic tensor. The model assumes a constant bulk modulus and has one fractional derivative of the deviatoric strain. The Jacobian of the constitutive equation with a fractional derivative is derived and implemented. The results from the subroutine are compared satisfactorily to results from the numerical Laplace inverse for a cubic element submitted to sinusoidal strains. / Finally, the different models are tested to represent the experimental behaviour of slewing beams made either of polymethylmethacrylate or steel covered by constrained viscoelastic layers. The classical models give generally a poor representation of the experimental behaviour, except for the Prony series. The fractional model give a representation as satisfactory as the ones obtained with the Prony series, but for a much higher CPU times due to the hereditary nature of the fractional derivative. It is therefore recommended to use Prony series models, unless the data to perform the parameter identification is limited. In that case, the fractional order model becomes interesting despite the higher demands on the CPU time.

An equal-order co-located Control Volume Finite Element Method (CVFEM) for the prediction of multidimensional, incompressible, viscous fluid flow problems has been formulated. CVFEMs provide the geometric flexibility traditionally associated with finite element methods. In addition, their control volume based formulation facilitates physically meaningful interpretation of the resulting discretization equations. / In the proposed CVFEM, the calculation domain is divided into three-node triangular and four-node tetrahedral finite elements in two- and three-dimensions, respectively. Each element is further subdivided in such a way that upon assembly of all elements, complete control volumes are formed about each node in the calculation domain. Interpolation functions for the dependent variables are prescribed in a manner that is consistent with the physical process they are intended to approximate. In this context, three different interpolation schemes of the convective flux across control volume surfaces are investigated, one of which guarantees positive contributions to the coefficients in the algebraic discretization equations. Appropriate conservation laws are imposed on the control volumes associated with the nodes. The resulting sets of integral conservation equations are then approximated by algebraic discretization equations, using the previously-mentioned interpolation functions. These nonlinear, coupled, algebraic equations are solved by a sequential solution procedure which incorporates Picard iterations. / The proposed method has been implemented into computer programs, and used to solve several test problems. These include convection-diffusion problems, and laminar and turbulent flow problems, in both two- and three-dimensions. The results demonstrate the ability of the proposed CVFEM to accurately solve the mathematical model used in this thesis. / Lastly, the CVFEM was used to predict flows similar to those found in film cooled gas turbine aerofoils. A complementary experimental program was designed and set up to investigate such flows. The numerical predictions were compared to the experimental observations of mean velocities, normal turbulent stresses and one component of the turbulent shear stress. These comparisons indicated that the high Reynolds number $k- epsilon$ turbulence model used in this thesis is unable to capture certain features of the flow. It appears as if a low-Reynolds number turbulence model, with appropriate modifications to account for streamline curvature and non-isotropy of the turbulence in the vicinity of the walls, would be better suited to the prediction of such flows. This new model can be easily incorporated into the CVFEM proposed in this thesis.

In this thesis, a unified formulation for the synthesis of cam mechanisms is presented, which allows the design of spatial, spherical and planar mechanisms within a single theoretical framework. This formulation has led to the design of novel cam mechanisms, as yet unknown. / The theory is then extended to the study of four-link cam mechanisms, where an intermediate element, namely, a roller, is placed between the cam and the follower. In general, the surface of the roller is shown to be a hyperboloid, which, for the planar case, becomes a regular cylinder, and, for the spherical case, a regular cone. The two theorems stated for three-link cam mechanisms are then extended to four-link mechanisms. A general formulation is presented for the study of the pressure angle in both three- and four-link cam mechanisms. With this formulation, the general expression for the pressure angle of spherical mechanisms is derived as a particular case of the general expression for spatial cam mechanisms. Moreover, the pressure-angle expression corresponding to planar cam mechanisms is derived as a particular case of spherical cam mechanisms. / The unified formulation is then applied to the synthesis of planar cam mechanisms via graphical methods. Here, an innovative technique is introduced, which consists of finding contact points of the cam profile, in contrast to traditional graphical techniques that employ cam envelopes, and, hence, are prone to inaccuracies. / Special attention is given to the synthesis of indexing cam mechanisms. In this context, a theorem is established for the determination of one of the design parameters that is used to avoid undercutting on the cam profile. A novel design of an indexing cam mechanism, called PRICAM, in which pure rolling and positive motion are achieved for planar and spherical mechanisms, is obtained using a combination of three- and four-link cam mechanisms. / The unified formulation is implemented in the software package USYCAMS, in which, with visualization aids, the user can design cam mechanisms of the three types and animate their motion, by providing the design parameters on-line. USYCAMS allowed the design of two versions of PRICAM, one planar and one spherical, and produced a database describing the contact surfaces, which served as input to the CNC machine tool used to cut actual prototypes of these mechanisms. (Abstract shortened by UMI.)

The quasi-steady assumption is frequently employed to derive the fluid dynamic forces on an oscillating body in fluidelastic stability analysis. Although it is well accepted, and physically reasonable, that this assumption is valid for high values of non-dimensional flow velocity, $U/fD,$ and invalid for low values, it is still uncertain how the error varies with $U/fD.$ In this thesis the applicable range of the quasi-steady assumption and the error of the assumption are investigated experimentally for three cases of two staggered circular cylinders in cross-flow. The applicable range is presented in terms of $U/fD,$ reduced frequency $fD/U$ or the ratio of the vortex-shedding frequency on the stationary model to the frequency of oscillation, $f sbsp{v}{s}/f.$ / Two quasi-steady models are studied, a basic model and a modified model. The basic model allows fluid force coefficients measured on a stationary body to be used to determine the fluidelastic forces on an oscillating body, provided that the body motion is correctly accounted for in the resultant velocity vector. The modified model takes the following into account also: (i) the magnitude and incidence of the flow velocity approaching the oscillating cylinder, and (ii) the time delay between cylinder displacement and the fluidelastic forces generated thereby. / Forced vibration measurements are conducted for the three cases. In each case one cylinder is forced to oscillate in the cross-flow direction and the other is fixed. The fluidelastic forces on the oscillating cylinder in the cross- and in-flow directions are measured over a wide range of $U/fD$ (roughly $5f/f sbsp{v}{s}>1.1$, where the quasi-steady assumption totally breaks down.

The work reported in this thesis mainly involved the development, implementation, and testing of mathematical models and numerical solution methods for computer simulations of temporally and spatially periodic, fully-developed, laminar and turbulent flows in rectangular interrupted-plate ducts. A relatively limited complementary experimental investigation was also undertaken to measure vortex-shedding frcquencies for turbulent flows in the ducts of interest. / The fluid flows considered in this work are characterized by the Reynolds number, Re, and four geometric parameters: The dimensionless streamwise length and half-thickness of the plates, L/H and $ delta /H$, respectively; the nondimensiond streamwise inter-plate spacing, s/H; and the aspect ratio of the rectangular duct cross-section, b/H. Here, H is the half-height of the ducts. In the numerical investigations, it was assumed that $b/H gg 1$ and the fluid flow is two-dimensional. / In the numerical investigation of laminar flows, six different values of the integer periodicity index, m, which refers to the number of geometrically similar modules included in the calculation domain were considered: m = 1, 2 3, 4, 5, and 6. The main findings of this investigation are the following: For $m>1$, multiple stable solutions are possible; the one-module ($m=1$) solution represents one of the possible stable solutions for the multiple-module ($m>1$) calculation domains; the magnitude of the maximum difference between the single- and multiple-module results of the Strouhal number, S, and the time-averaged friction factor, f f, can be as high as 29.2% and 13.9%, respectively; and the critical Reynolds number for the onset of temporally and spatially periodic oscillations increases as the nondimensional plate thickness, $ delta /H$, decreases. It was also found in this investigation that for the thinnest plates considered, $ delta /H$ = 0.12, the numerical solution of the steady-state formulation of the problem provides good estimates of the time-averaged friction factor, f f, and the average nondimensional viscous shear stress, $ tau sbsp{w}{*}$, on the top and bottom surfaces of the plates. However, for $ delta /H$ = 0.20 and 0.32, the unsteady simulations are necessary to obtain accurate values of ff and $ overline{ tau sp{*}} sb{w}$. / It was concluded that at present, a combination of complementary experimental and numerical investigations with fine-tuning of some of the closure coefficients in relatively simple turbulence models such as low-Reynolds number two-equation eddy-viscosity models, may be the most effective approach to the prediction of temporally and spatially periodic, fully-developed turbulent flows in interrupted-surface geometries. (Abstract shortened by UMI.)

A theory to calculate the total force and torque acting on a long slender body at rest in a fluid undergoing a uniform viscous flow was developed by R. E. Khayat and R. G. Cox. The present experimental investigation was conducted to study the applicability of this theory. The translational and the rotational velocity of eight different slender rods, each sedimenting at various orientations in five different silicon oils were determined by taking single-frame multiple-image photographs. The data from the films were extracted using a two dimensional measuring microscope, and numerical calculations were carried out to calculate the experimental drag, lift and torque. The experimental findings were compared with the predictions from the theory for a range of Reynolds number R, based on cylinders cross sectional dimension $(0.00005 le R le 23.3).$ / The measured experimental and the theoretical drag force were consistent to within the order of accuracy of the experiments. This drag force was an increasing function of the Reynolds number for all orientations, with the maximum drag force occurring for horizontal cylinders. The experimental lift force also showed a fair agreement with the theoretical predictions for the range of Reynolds number $0.00005 le R le 0.2$ at all cylinder orientations. In the range of $0.00005 le R le 0.2,$ the cylinder observed the highest lift force at $ theta = pi/4$ where $ theta$ is the angle of cylinder's axis makes with the free stream direction. The comparison of the experimental and the theoretical rotational velocity showed good agreement for $0.00005 le R le 0.001,$ but relatively poor agreement for $0.001 le R le 23.3.$