Experiments in the study of transient aerodynamics typically require pressure measurements with a high spatial and temporal resolution. Existing commercial pressure transducers are expensive and they provide a spatial resolution only on the order of millimetres. The full bandwidth of commercial devices (which extends to around 200 kHz) can only be utilised by exposing the transducer to the flow environment with very little thermal or mechanical protection. If insufficient protection is provided, the expensive commercial devices are likely to be damaged. Inexpensive pressure sensors based on extrinsic Fabry-Perot fibre optic interferometry are capable of measurement with a high spatial and temporal resolution. Thermal protection or isolation for these sensors is still required, but they can be exposed directly to the flow if the sensors are disposable (low cost). Excessive thermal or mechanical protection is not required for these sensors because the damaging heat transfer and particle impacts that may occur in transient aerodynamic facilities generally occur after the useful test flow. In this dissertation, a variety of construction techniques for diaphragm-based Fabry-Perot fibre optic pressure sensors were investigated and the advantages and disadvantages of all techniques are compared. The results indicate that using a zirconia ferrule as the substrate, a liquid adhesive as the bonding layer, and a polished copper foil as the diaphragm provide the best results. It is demonstrated that a spatial resolution on the order of 0.1 mm and a bandwidth to more than 100 kHz can be achieved with such constructions. A variety of problems such as hysteresis, response irregularity, low visibility and sensor non-repeatability were observed. By using a thinner bonding layer, a larger bonding area, longer cavity length, increased calibration period, and applying load cycling to the diaphragm, the hysteresis was minimized. Sensor response irregularity was also minimized using a polished diaphragm. Visibility increased to about 90% using active control of the cavity length during the construction process. Non-repeatability was found to be a consequence of adhesive viscoelasticity and this effect was minimized using a thin layer of adhesive to bond the diaphragm to the substrate. Due to the effects of adhesive viscoelasticity, the pressure sensors indicate an error of up to 10% of mean value for the reflected shock pressure. This error could not be further reduced in the current sensors configuration. Some new configurations are proposed to decrease the effect of sensor non-repeatability. The effect of pretensioning the diaphragm was investigated analytically but the results do not indicate any considerable advantage for the levels of pretension likely to be achieved in practice. However, the results do indicate that pretension effects caused by an environmental temperature change can damage the sensor during storage. The effect of the initial diaphragm deflection on the sensor performance and temperature sensitivity was modelled and the results show that an initial diaphragm deflection can improve the sensor performance. The effect of the thermal isolation layer on the sensor performance was also investigated and the results show that for a shock tube diaphragm bursting pressure ratio up to 5.7, heat transfer does not contribute to sensor errors for the first millisecond after shock reflection. However, it was found that the use of a thin layer of low viscosity grease can protect the sensor for about 20 ms while only decreasing its natural frequency by typically 17%. The grease layer was also found to decrease the settling time of a low damping ratio sensor by 40%. The sensor was successfully employed to identify an acoustic disturbance in a shock tube.

Thesis (M. Phil.)--University of Hong Kong, 2009. / Includes bibliographical references. Also available in print.

Thesis (M. Phil.)--University of Hong Kong, 2002. / Includes bibliographical references (leaves 117-121).

In recent years, studying soft errors has become an issue of greater importance. There have been many methods developed that estimate the Soft Error Rate. Those methods are either deterministic or statistical. The proposed deterministic model aims to improve Soft Error Rate estimation by accurately approximating the generated pulse and all subsequent pulses. The generated pulse is approximated by a piecewise function consisting of two Weibull cumulative distribution functions. This method is an improvement over existing methods as it offers high accuracy while requiring less pre-characterization. The proposed algorithm reduces pre-characterization by allowing the beta Weibull parameter to be calculated during runtime using gate parameters such as the gate delay.

Low dropout regulators (LDOs) are important components for power management in modern integrated circuits. With the continued scaling down of power supply voltage, digital LDOs have become a more attractive design choice since they avoid the difficulty of designing high-gain amplifiers with low voltage. This thesis investigates techniques for both modeling and enhancement of digital LDO transient response. It discusses the importance of the resistance in the output stage of an LDO, and proposes a simulation model for examining LDO transient response. In addition, the thesis studies circuit techniques to improve LDO transient response. Different LDO circuits are implemented and compared in this study.

L'auteur n'a pas fourni de résumé en français / L'auteur n'a pas fourni de résumé en anglais

This thesis is based on an ethnographic study of the inhabitants of Whiddy Island, and focuses on the change from one form of societal organisation to another on this island. The thesis is not an ethnography proper, but an attempt to link the local perceptions of change and the changes in the islanders' daily lives, to the wider political economy. Throughout the course of the study my original intention of exploring the tension between technology and community was replaced with the wider hypothesis that there is tension between modernity and community. Technology was revealed as both a product and producer of modernity, and modern state capitalist societies as the antonym not the synonym of community. The 40 remaining islanders represent the last of the transient phase in which community disappears and is replaced by society. The changes in the daily lives of the islanders were not total nor revolutionary. Rather the products of modernity - both policies and artefacts, were absorbed into the islanders' daily lives, and once absorbed the products of modernity promoted modernity in the daily lives of those using them. Modernity is thus a circular process, yet it settled on the island in layers. Each layer produced a new set of paradoxes and reformed the old practices and the old ideology to fit the new setting. The settlement of modernity culminated in the replacement of community members with state citizens. By focusing on the interrelationship and dialogue between modernity, the state and the citizen the processes by which modernity settled on this small island are revealed. It settled both as a result of the direct intervention of state policies on education, emigration and employment, and as a result of local decisions to embrace mechanised transport, domestic technologies and the mass media. By accepting the policies and the artefacts of modernity, the islanders were prohibited from resisting their transformation from community members to state citizens. The island citizen, like all citizens to-day, has a direct dialogue with, and relationship to modernity, and an indirect one mediated by the state.

The effect of surface roughness on jet impingement heat transfer was investigated in this research. A numerical analysis was conducted for free surface slot jet impinging normally onto a heated plate. Six different geometries and three different plate materials were investigated. The cooling fluid used for the analysis was water, and the flow was laminar with a range of Reynolds number (Re) from 500 to 1000. Temperature distribution, local and average heat transfer coefficient, and local and average Nusselt number were presented for each case. The steady state heat transfer results show that the increase in Reynolds number (Re) increases the local heat transfer coefficient and the local Nusselt number. Impinging the jet nozzle directly onto a step has a better heat transfer enhancement than impinging the jet nozzle in between steps. Materials with low thermal conductivity exhibit large variation in temperature along the solid-fluid interface. The variations of the interface temperature become smaller between all cases when applying the isothermal boundary condition. The transient heat transfer results show that the temperature of the interface increases with time until steady state condition is met. Materials with high thermal diffusivity reach the steady state condition with less time. The increase in surface roughness increases the time required to reach the steady state condition. The highest rates of heat transfer were found at locations where no fluid recirculation occurs. It takes less time to reach steady state condition when applying the isothermal boundary condition at the bottom surface of the plate.

No description available.

Simulation tools capable of analyzing electromagnetic (EM) field/wave interactions on complex penetrable scatterers have applications in various areas of engineering ranging from the design of integrated antennas to the subsurface imaging. EM simulation tools operating in the time domain can be formulated to directly solve the Maxwell equations or the integral equations obtained by enforcing fundamental field relations or boundary conditions. Time domain integral equation (TDIE) solvers offer several benefits over differential equation solvers: They require smaller number discretization elements/sampling points (both in space and time). Despite the advantages, TDIE solvers suffer from increased computational cost, stability issues of the time-marching algorithms, and limited applicability to complex scatterers. This thesis is focused on addressing the last two issues associated with time domain volume integral equation (TD-VIE) solvers, as the issue of increased computational cost has been addressed by recently developed acceleration methods. More specifically, four new closely-related, but different marching on-in-time (MOT) algorithms are formulated and implemented to solve the time domain electric and magnetic field volume integral equations (TD-EFVIE and TD-MFVIE). The first algorithm solves the TD-EFVIE to analyze EM wave interactions on high-contrast dielectric scatterers. The stability of this MOT scheme is ensured by using two-sided approximate prolate spherical wave (APSW) functions to discretize the time dependence of the unknown current density as well as an extrapolation scheme to restore the causality of matrix system resulting from this discretization. The second MOT scheme solves the TDMFVIE to analyze EM wave interactions on dielectric scatterers. The TD-MFVIE is cast in the form of an ordinary differential equation (ODE) and the unknown magnetic field is expanded using spatial basis functions. The time-dependent coefficients of this expansion are found by integrating the resulting ODE system using a linear multistep method. The third method is formulated and implemented to analyze EM wave interactions on scatterers with Kerr nonlinearity. The former scheme integrates in time a coupled of system of the TD-EFVIE and the nonlinear constitutive relation, which is cast in the form of an ODE system, for the expansion coefficients of the electric field and flux using a linear multistep method. The last method described in this thesis is developed to analyze EM wave interactions on ferrite scatterers.