A special type of singular Stewart-Gough Platform exhibits continuous motion of one or more degrees of freedom, while all of its six or more legs are locked at fixed length. The equation describing point motion is introduced and then applied to parallel platforms. Line geometric analysis shows that all legs remain in a continuously varying linear complex, linear congruence, or quadric. Base and platform joints are restricted to lie on two congruent, skewed hyperbolids of one sheet. One of these constitutes those points on the rigid body of the platform that move on spheres, the other, those on the fixed base which constitute the sphere centres. An equivalent mechanism is developed and used to visualize the design parameters of the motion. The selection, as platform joints, of points on a line generator results in a two degree of freedom platform. Specific designs of such overconstrained mechanisms are investigated and illustrated. Line geometry is used, to explain the special relationship among the platform legs. Numerical examples are given and illustrated.

This thesis describes experiments on granular flow from a flat-bottomed bin with an attached standpipe. The experiments focused on three key aspects of standpipe-flow: (i) the effect of length-to-diameter ratio (L/D) on the mass flow rate and interstitial air pressure distribution, (ii) the effect of surcharge on the flow rate, and (iii) the effects of apparatus grounding and electrostatic charges on flow characteristics. Two materials (coarse sand and fine glass beads) were tested in both the loose and dense flow regimes. In general, experimental data correlated well with previous experimental work, but not with theoretical predictions. Theoretical predictions of the pressure distributions agreed qualitatively with the experimental results in the loose flow regime, but not in the dense flow regime. Mass flow rate theories for the loose flow regime are qualitatively and quantitatively inaccurate when compared with experimental results. The effect of surcharge level on the mass flow rate was found to be insignificant. The transition between loose and dense flow regimes was studied with respect to both static electricity and standpipe air pressure. Experiments suggest that static electricity may play a key role in the flow dynamics of standpipe systems.

Robots with four legs offer a good tradeoff between stability, load carrying capacity, and mechanical complexity when compared to bipeds and hexapods. In order to achieve the best mobility, speed, and energy efficiency, dynamic walking and running operation is preferable to static gaits. This requires leg compliance to reduce impact forces and energy consumption. / At the Ambulatory Robotics Laboratory, (ARL), we have pursued an agenda of low mechanical complexity in our Scout I and II robots, in order to decrease cost and increase reliability. Research previously undertaken at the ARL group has accomplished walking and stair climbing with Scout I and walking with stiff legs with Scout II. In this thesis, we demonstrate that Scout II, with only an additional compliant prismatic joint per leg, is able to bound. We show that dynamic running is possible with very simple control strategies. Open loop control, where switching torque values at the hip during support or flight phases results in a stable bounding gait. We also investigate more elaborate controllers that control forward speeds. The bound controllers were first developed and validated in simulation. These strategies were then implemented on the Scout II robot yielding successful running at speeds of up to 1.2 m/s.

Submarine cable systems are a vital and expanding part of the world wide telecommunication network. The systems are laid in the ocean using a cable ship. During the deployment of a submarine cable, the cable ship has only limited information about the behaviour of the cable once it leaves the vessel and begins its descent to the seabed. The present thesis studies the cable dynamics and control during laying operations. / The cable dynamics is modeled using the continuum approach. Using the model developed, the cable configuration and tension are obtained for the two dimensional steady state and transient motions, when different parameters such as the ship velocity, hydrodynamic constant, and current direction, are varied. Also, the out of plane angle of inclination and corresponding maximum displacement for the three dimensional steady state case are determined. / Then the possibility of controlling the cable oscillations by varying the payout rate is explored. In the case of the controlled response the results show that, in principle, the deviation of the in plane angle can be effectively reduced to zero, but the value of the input payout rate needed is unrealistic in practice.

The effect of cool flame partial oxidation on the detonation sensitivity of a hydrocarbon fuel was investigated experimentally. The detonation sensitivity was quantified by measuring the run-up distance required for a deflagration to transit to a detonation wave (DDT) in a rough tube. Fuel rich pentane-oxygen mixtures at subatmospheric initial pressures were studied. Subsequent to the injection of the mixture into the heated detonation tube, the mixture underwent cool flame oxidation after a well controlled delay time, as it was determined by the temperature of the tube. Typical delays ranged from 0.7 to 2 seconds (depending on temperature) and were reproducible to within one hundred milliseconds. This permitted to spark-ignite the mixture in the detonation tube at various stages of its cool flame process. The results show that increasing mixture temperature from room temperature to values below the cool flame region (below 250°C) resulted in an increase in run-up distance. However, as the mixture begins to undergo cool flame oxidization a significant reduction in the run-up distance was obtained (as large as 50%). The sensitization effect was found to occur only at the initial stage of the cool flame oxidation reaction. If the mixture is ignited at times long after the onset of cool flame, the mixture was found to be desensitized and the run-up distance increased. The sensitizing effect of the cool flame partial oxidation may be attributed to the presence of peroxides and free radicals associated with the initial cool flame process. However these radical species are consumed as the cool flame reaction proceeds and the mixture becomes insensitive again.

Currently there is no standard whole-field methodology to measure the transient out-of-plane deformation of a surface under dynamic events. In this work, a whole-field high-speed optical surface measurement system has been developed, which yields absolute Cartesian coordinates (xyz) as measurement results. A high-speed digital camera coupled with the grating projection and Fourier transform is used to measure 3D surface deformations with the maximum sampling rate up to 1000 Hz and the exposure time of each frame up to 50 musec. A calibration technique has been utilized to convert the direct measurable values---phase, phi, and image indices (i, j)---into 3D Cartesian coordinates (xyz). In addition, all Cartesian coordinates are with respect to a fixed coordinate system, which is virtually associated with the sensor head. Therefore, the rigid body movements (rotation and translation) of the surface can be detected. The measurement accuracy +/-50 mum is verified with several tests by using standard objects. / The dynamic surface deformation during polymer membrane inflation tests is investigated by using the proposed optical measurement system. Effects of various factors, such as the temperature and the airflow rate, are studied. Regarding the characterization of material constants by using the bubble inflation technique, the potential problems, which are caused by ignoring the nonuniformities of temperature distribution and the thermo-warping, have been studied. / A finite-element model, which can account for the nonuniformities of temperature distribution and the thermo-warping, has been employed to simulate the inflation deformation. A more accurate way to characterize the material constants is accomplished by combining the optical measurement system and a finite-element model that can account for the nonuniformities of the temperature distribution and the thermo-warping.

A computational study is carried out to investigate the heat, mass and momentum transfer phenomena in a novel two-dimensional opposing jet gas-particle contactor. Initially, the confined laminar opposing jet configuration is examined by solving the unsteady and steady conservation equations. Both equal and unequal momentum jets impinging normally against each other are considered. The thermal boundary conditions tested are: uniform wall temperature and insulated walls. Predicted velocity fields, pressure drop, and heat transfer coefficient distributions are compared with corresponding values for laminar flow between parallel plates. Mixing of the jets in such flows is discussed in terms of the variation in temperature profiles in the exit channel. / An evaluation of the predictions of different high- and low-Reynolds number turbulence models for turbulent flow and heat transfer in confined impinging and opposing jet configurations is carried out. The heat transfer features of these systems are compared on the basis of equal mass flow rate in the two systems. A parametric study for flow and heat transfer in the turbulent opposing jet case is carried out. / The motion and drying of single particles moving in two dimensional opposing jet contactor in the turbulent regime are investigated. The effects of key parameters such as particle size, release position in the inlet jet, jet Reynolds number, and geometry of the contactor on particle residence time and particle moisture content are examined, assuming a power-law model for the falling rate drying kinetics. / A two-phase flow model is developed based on a modified stochastic particle-trajectory model. Both turbulence modulation and particle dispersion effects are incorporated in the code. The effect of turbulence intensity on particle-gas heat transfer is included via a new model. A modified algorithm is introduced for the solution of the governing equations. / The proposed mathematical model and solution algorithm are applied to selected test problems, e.g. dilute gas-particle flow in a channel. The predicted results compared favorably with results of independent investigations. / The two-phase code is employed for a parametric study of the flow and drying characteristics of wet particles in the proposed opposing jet contactor using superheated steam as the drying agent. Predictions are made for various degrees of superheat and operating pressure. The effects of particle size and jet Reynolds numbers are also investigated. The particle residence time and moisture content distributions are predicted and a new method is suggested for an effective use of computer simulation results in the design and analysis of convective dispersion dryers for particulate materials.

This thesis deals with the investigation concerning a predictive compensation method which corrects for the error on the diameter caused by the deflection when machining a workpiece on a lathe. / The deflection model which was formulated to predict the error on the diameter considers the smaller emerging diameter formed by the depth of cut as the tool tip moves along the workpiece, thus improving the accuracy with which the deflection can be predicted. The model also recognizes the fact that the total deflection undergoes an equilibrium search causing the final deflection to reach a smaller value than that which is obtained when using the methods described in literature. Furthermore, the way in which the workpiece is commonly supported in industrial practice, which is having one end rigidly clamped in a chuck at the spindle while the other end is supported by its center at the tailstock, is included in the deflection analysis. / The compensation method can be applied not only to uniform cylindrical bars but also to general shaped workpieces. However the latter must first be partitioned into segments forming a multidiameter shaft for the deflection analysis. Implementation of this predictive compensation method introduces the possibility of obtaining single pass machining with final pass accuracy.

Over the last decade, there has been considerable interest and success in the formulation of co-located equal-order Finite Volume Methods (FVMs) and Control-Volume Finite Element Methods (CVFEMs) for the prediction of fluid flow and heat transfer. This thesis is concerned with an evaluation and enhancement of some aspects of recent co-located equal-order FVMs and CVFEMs. In particular, the goals of this thesis are the following: (1) formulation and computer implementation of a co-located equal-order FVM that facilitates the evaluation and enhancement tasks; (2) evaluation and enhancement of iterative sequential and coupled-equation solvers; and (3) comparative evaluation of a recently proposed mass-weighted skew upwind scheme (MAW) against five well-established schemes. / The proposed equal-order co-located FVM is formulated for the prediction of steady, two-dimensional, incompressible, viscous fluid flow in planar rectangular domains. This FVM deals directly with the velocity components and pressure, or primitive variables. The rectangular domains are discretized using structured line-by-line rectilinear grids, and rectangular control volumes are constructed around each grid point. All dependent variables are co-located or stored at the same grid points, and interpolated on the same rectangular elements in an equal-order formulation. A computer program incorporating the proposed FVM has been developed and tested successfully. / On the basis of an evaluation of some available iterative solvers, two improved algorithms are proposed in this work: (1) Enhanced Sequential Solution Algorithm (ESSA); and (2) Sequential Variable Adjustment (SEVA) algorithm. Test results obtained with these algorithms are very encouraging, particularly with ESSA. The comparative evaluation of the MAW scheme shows that its performance is comparable to the skew upwind difference scheme (SUDS) of Raithby (89).

The aim of the present research is to develop a theoretical model that could predict the trailing-edge separation over a circular-arc aerofoil and its effect on the pressure distribution, and on the lift. The model uses sources, in a potential flow, to simulate the effect of the separated region on the pressure distribution. An integral method, turbulent boundary-layer development, combined with the irrotational pressure distribution on the circular-arc predicts the point of separation. To make sure that trailing-edge separation is present alone and mostly to avoid any difficulties in predicting the boundary-layer flow after reattachment, the aerofoil is studied at design incidence. / Two different circular-arc aerofoils were tested; one of 10% and one of 18% camber. The calculation for the 10% aerofoil did not predict any separation at design incidence, which was confirmed experimentally. The theoretical model predicted the right position of separation for the 18% aerofoil. It also provided a good simulation of the pressure distribution, including the right value for the base pressure and a good prediction of the lift. It follows that the pressure distribution when integrated, gives a good estimate of the pressure or form drag. (Abstract shortened by UMI.)