Ion plating is a fairly new and modern technique for obtaining very adhesive films on any kind of surface and substrate material. The development of electron gun evaporation sources created the possibility to deposit, with high rates, materials having high melting points like : Molybdenum, Tungsten and even Carbon. A part of this work attempts to establish the kind of particles which are involved in ion plating, especially the energetic neutrals that have been considered not to play a major role in the coating process. For this purpose, the ions are separated from the neutrals and their energy spectrums determined using theoretical and experimental methods. The influence of process parameters upon the energies of the ions and neutrals is also studied. Another part of this work is a parametric study of the nucleation of ion plated thin films as compared to the straight vacuum evaporated ones. The film-substrate interface of ion plated films is also analysed and measured attempting to clarify the processes that lead to the formation of a deep grossed interface. Finally a parametric study is carried out with the aim of producing improved film structures in a controlled manner by varying the main process parameters such as : pressure, bias voltage and ion current. The substrate temperature is also considered. A physical model for removing the columnar structure is produced snd reported.

Abrasive waterjet (AWJ) cutting is one of the most promising fast emerging non-traditional cutting technologies. It is highly competitive for machining difficult-to-cut materials like ceramics, composites and titanium alloys as compared to other nonconventional processes (e.g. laser, EDM) which are either technologically inappropriate or fail to be cost-effective. However, at the moment most of the usage of the AWJ machining lies in the area of the through cutting applications and to perform controlled depth cutting (milling) is still at craftsmanship level. This is due to the facts that: (i) AWJ machining is based on employing a jet plume as a "soft body" tool, the footprint of which not only depends on the jet energy parameters (e.g. pressure, abrasive mass flow rate, etc) but also on the jet kinematic parameters (e.g. jet traverse speed) which make controlling of the jet penetration depth very difficult; (ii) there is absence of the appropriate and reliable models that can simulate and predict the AWJ milled footprints and this is one of the major obstructions constraining the use of the AWJ milling applications. The aim of this thesis is to develop accurate models for predicting the A WJ milled footprints. The workpiece material considered is a titanium based superalloy (Ti-6Al- 4V) which is extensively used in the aerospace and medical industry. Two modelling approaches; finite element (FE) modelling and mathematical modelling are presented in this work. Considerable numbers of experiments are conducted to generate the data for validating the results from the models. The models presented in the current study are closer to the real life conditions occurring during the A WJ machining as compared to the state of the art in modelling of AWJ machining. Regarding the FE modelling, the abrasive particles (i.e. garnet) are modeled as elastic with a tensile failure criterion with various non-spherical shapes (rhombic, triangular and trapezoidal) and sharp cutting edges in contrast to the usual approach of assuming them as rigid spherical particles. The effects of mass flow rate of the abrasive particles, traverse speed of the AWJ plume across the workpiece and Gaussian spatial distribution of the abrasive particles in the jet plume are also incorporated in the FE model. The FE model is developed to an extent that it can simulate the footprints as a result of overlapping passes of the AWJ. The simulated jet footprints from the FE models are in good agreement (maximum errors ≤ 15%) with the experimental results. From the mathematical modelling point of view, a model is developed that can accurately predict the AWJ milled footprints with root-mean-squared errors less than 9%. The model takes into account the effects of jet incidence angles, traverse speeds and arbitrarily-moving jet-paths within the target surface. The model is computationally inexpensive and can be used for real time predictions of footprints during CNC machining. The current study provides the reliable models that can be employed for accurate prediction of the abrasive waterjet milled footprints at various process parameters which is a necessary step towards the exploitation of the A WJ machining for controlled depth cutting applications and its automation.

This thesis is written to give an overview of the most important types of calculation methods for the analysis of y-type Stirling engines found in the last 60 years. Simple methods like the ideal process calculation and the 0th order analysis found by Beale and West are given to describe the process steps and to get a first reference value of performance and efficiency. Higher order calculations like the Schmidt analysis (1st order) and 2nd order methods for the ideal adiabatic and quasi steady flow models are described in detail and optimised for the y-type Stirling engine. With a generated quasi steady computer program code parameter variation is used to obtain an impression of the dependency of performance and efficiency on varying geometry data and boundary conditions. In addition to these models the heat exchanger sections heater, cooler and regenerator are analysed in depth with the CFD program ANSYS CFX. To compare the results of the theoretical analysis to measured ones three experimental engines are used. Different process values are determined on a biomass fired CHP Stirling system and on a solar Dish / Stirling system. On a Stirling engine test bench some of the parameter variations of the quasi steady program are repeated in experiments for comparison. These engines are modified in ways to make them run properly and to improve durability. The behaviour of the y-type Stirling engine is analysed in detail both in experiments and theoretically: this is felt to be unique. With the modified quasi steady flow model a method is found that is able to predict the process performance with a higher accuracy than it can be done with any other calculation method. This method can easily be modified to fit any other type of Stirling engine.

A novel method for calculating the linear fluid-structure interaction of a cantilevered flexible surface centrally positioned in an ideal channel flow, incorporating the effects of vorticity shed downstream, is described. The perturbation pressure is modelled using a linearised boundary-element method. The flexible surface deflection is modelled using linearised one-dimensional beam theory. The shed vorticity is modelled using a linearised discrete vortex method. The computational model can therefore be used to conduct numerical experiments where no presupposition of the flexible surface deflection is made. This linear model can accurately capture the onset of instability in this fluid-structure system. The flexible surface is infinitely thin; the upper and lower sides of the surface can therefore be considered stream lines of the flow, with a step jump in pressure between them across the surface. The discontinuity of tangential velocity across the flexible surface generates lift. The flexible surface is therefore modelled by a distribution of vortex singularities with a Kutta condition applied at the surface’s trailing edge. The individual models of the flexible surface and the fluid velocity and vorticity, together with the action of the individual hydrodynamic pressure components created when the models are combined to form a single unsteady model, are validated via a series of numerical experiments and by quantitative comparison with an appropriate, previously developed computational model. Unique, highly detailed investigations into the ideal fluid-structure phenomena observed in numerical experiments conducted over a wide range of mass ratio and inlet velocity are documented. For the first time, detailed numerical investigation of the effect on this fluid-structure interaction of channel walls, a rigid central surface (upstream and adjacent to the flexible surface), unsteady mean flow, the variation of stiffness and damping properties along the flexible surface and the vorticity shed at the trailing edge of the flexible surface have been quantified. Calculations of the critical velocity show good correlation with other published work and examples of the possible application of the unsteady model to different physical fluid-structure phenomena are outlined. Of central importance is the application of the unsteady model to the investigation of the human snoring phenomenon. Further insight into the operation of two types of snore is made and a new type of snore is discovered, incorporating the effects of inhalation. The numerical experiments demonstrate that the location (on the flexible surface) of the destabilising phase shift between the flexible surface velocity and fluid pressure leading to instability change drastically for a small shift in mass ratio. Coupled with knowledge of further snore mechanisms from other published work, these results show the uniqueness of treatment required to provide effective surgical treatment to individual patients suffering from snoring; furthermore, this highlights the need for more realistic fluid-structure models to be created.

At a high level of abstraction, the brain is a system for analysing sensory information, and responding appropriately. That information is encoded and stored in the millions of neural circuits that comprise the brain. Deciphering this code is essential to understanding how memories are implemented in physiologically normal brain tissue, and to inferring the nature of some neurological disorders affecting memory such as Alzheimer’s disease, in which the neural encoding is aberrant or unsuccessful. One approach to this problem is to reduce the complexity of the brain functionality to three elements: stimuli, response, and reinforcement. The electrical activity of individual neurons can be recorded with electrodes, capturing the pathways of signal propagation in a network of cells. Individual neurons can be also induced to reliably respond to electrical or optical stimuli, so that they initiate, relay, or even block a signal. If the stimuli to a finite network of cells can be made heterogeneous so that only a sub-population of cells is targeted, then the network can be trained to react in a repeatable way to a given stimulus, testing the concept that the higher order functions of the brain can emerge from a simple set of underlying computational rules. Training however requires a mechanism for reinforcing only some of the possible pathways, in synchrony with stimuli and in response to the recorded network activity. In the intact brain, this mechanism is pharmacological: a neuromodulator such as dopamine is released throughout the brain, but as it only coincides with some but not all neuronal activity, the reinforcement is temporally selective. The key task of this project is to emulate this selective neuromodulator reinforcement in vitro in a finite neuronal network. The project must also provide capacity for heterogeneous stimulation and individual cell recording, which can be coordinated with the reinforcement under computer control. The strategy used was to develop microscale chambers to house a small network of cultured neurons. The chambers were integrated with existing cell recording and stimulating technologies, so that specific connections between neurons could be both monitored and induced. Neuronal cultures of a few hundred cells were successfully grown in microchannels, on substrates capable of recording their electrical activity. Thus it was possible to create a small cultured network in which complete network activity could be detected, subject to a sufficiently precise recording technique. Additionally, a fluid-handling system was developed in order to emulate the continual replenishment of nutrients and soluble gases that are essential to cell survival. The system is intended to deliver soluble chemicals that modulate neuronal activity, on a timescale that is consistent with neuromodulator delivery in the body. The fluid handling system comprises a set of pressure driven pumps under automated computer control. This system has the capacity to deliver neuromodulator in solution with high spatiotemporal precision. The ability to reliably deliver and wash off precise volumes of drugs in a matter of seconds, with no dilution of the intended concentration, will be of great benefit to researchers investigating the response of various cell types to different agonists.

The behaviour of trains used at high speeds for transporting men along mine railways is examined using a number of mathematical models. These models predict the responses of different rail vehicles to typical irregularities in the track, and are used to examine the guidance, the ride, and the likelihood of derailment of the main classes of manriding trains used in British coal mines. The outcome of the modelling compares favourably with the results of tests carried out on actual vehicles. The investigation has shown that the safe speed at which trains may operate is ultimately restricted by the condition of the track, but changes in the design of the vehicles can lead to an improved performance. The use of conventional wheelsets, for example, can reduce flange wear significantly and so allows higher speeds to be reached. Guidelines for the design of vehicles intended for high speed use are included in this thesis.

Gas turbine compressor components manufactured from nickel base alloys have traditionally been precision die forged in sequential thermo-mechanical processing operations in order to achieve the desired geometry and mechanical properties. Metal Injection Moulding (MIM) is a competing three dimensional forming technology with proven applications in both the automotive and medical industries for producing industrial quantities of small, net shaped components. To date, the Metal Injection Moulding process has had limited exposure as a manufacturing process for gas turbine compressor components. The aim of this research thesis is to establish if the Metal Injection Moulding process can be used to manufacture compressor components of equivalent mechanical properties to those manufactured by conventional processing methods. In order to achieve this aim a rigorous program of metallurgical testing and analysis has been developed. The objectives of this program focus on determining the key material properties from each of the competing manufacturing processes. The methodology used to assess the merits of each process was based upon comparative back to back testing trials using both representative components and material test bars. The test results demonstrate that while the mechanical properties of the Injection Moulded 718 alloy can be improved by subsequent thermo-mechanical processing, there remains however a significant deficit in the strength, ductility and creep properties compared to the wrought 718 alloy datum results. The recommendations which have been made as a result of this research focus on improvements to the condition of supply of the powdered 718 alloy and to the controls associated with the Metal Injection Moulding process in order to minimise process variation. Standardisation of the mechanical testing methodology including the test piece geometry is also considered necessary in order to achieve a more meaningful comparison to published historical test data and allow wider industry corroboration of test results conducted in accordance with aerospace standards.

This thesis investigates the potential of using wind turbine to offset electricity demand for dwellings or public building. This work involves onshore small and large wind turbine implementation considering the suitability of the location to machine size, starting with wind resource assessment of a candidate site depending on reliable wind data. The present research can be divided into three main parts: modelling and monitoring of small wind turbine performance in built environment using detailed data which was measured on site, measuring longterm hourly data for the design of wind energy systems, and then comparing that annual energy output against four-second and minute by minute data. The third 'part presents a novel statistical tool developed to evaluate relative performance and overall accuracy of wind speed frequency distribution functions. An exploration of the potential for using hourly- as opposed to minute-by-minute data for the utilization of large wind turbines was undertaken as the former set is much more widely available for a larger number of locations within the developing world. It was found that the difference between the annual energy outputs from the latter two data sets was in close agreement with only small differences. The results thus obtained can have significant effect. on the capital cost related to purchase of data, since minute by minute data may be up to 60 times more expensive than hourly data. Actual power curve was experimentally obtained for Zephyr Dolphin micro wind turbine, which was then compared to manufacturer's reported performance; this was done by using four-second data for two complete years. Significant differences were found between the two curves. On-site measured performance of mentioned wind turbine was found to be similar for other reported urban locations. In each case the measured output was only a sixth of the acclaimed output of 2 MWh/annum.Urban wind energy potential for Merchiston site in Edinburgh was investigated. The results are presented in the form of average wind speed, wind roses, and density distribution functions. The effect of sampling interval on wind energy production was also analysed. Finally local spatial variations of wind speed were also studied for the City of Edinburgh.

Turbulent forced convective heat transfer is often encountered in engineering flows. A method of improving the heat transfer from a smooth surface is to add ribs and generate some turbulence. The precise nature and spacing of these ribs can have a considerable effect on the heat transferred. In this investigation the novel method of holographic interferometry is used to study the thermal fields over numerous two-dimensional ribbed geometries and to ascertain the Nusselt number distribution around them. Pull field information is provided as the fringe pattern generated is essentially an isotherm contour map of the flow situation. Hence, this makes it ideally suited to verify the theoretical solutions obtained from Large Eddy Simulations (LES) and Finite Element predictions. This latter predictive technique is used in this investigation and the solutions obtained are compared with the experimental results. Initially, a smooth surface geometry was investigated to verify the accuracy of the experimental technique. Excellent results were achieved, but a necessity to have momentum field information was identified. Ribbed geometries with a pitch to height ratio of 7.2 : I were then studied. Double exposure and real-time techniques enabled both detailed thermal measurements to be made and any time-dependency of the field to be identified. Flow rates up to nuclear reactor conditions (e+ = 600) were studied and typical interferograms are illustrated. A Reynolds number dependency for the heat transfer distribution for a rounded-rib geometry was identified while the Bquare-ribbed geometry distribution was observed to be Reynolds number independent. The addition of an insulating deposit to the surfaces is also studied. It led to smaller peak heat transfer rates and larger surface temperature variations. For large thermal gradients, usually at high flow rates, a ray crossing regime was identified. This led to a limitation of the technique as no information in this area could be extracted. However, a boundary layer extrapolation from the identifiable regions to the wall eradicates this limitation, but is only possible if momentum field information is available. Hence, this problem did not act as a limit to the smooth surface or square-ribbed geometry but did for all others. Numerical simulations of the flow were undertaken using the k-e and q-f models employed in the finite element code FEAT. Because a low flow rate was modelled the conventional wall functions were not considered appropriate and hence a low Reynolds number model was used. Incorrect modelling of the length scales in the near wall region using this model led to errors in the thermal field predictions. Hence, only a qualitative comparison with the experimental results is undertaken and a recommendation for improved modelling is proposed based on this comparison.

The performance of conical diffusers with axial and swirling flew has been considered. As a necessary starting point the various criteria used for defining performance have been reviewed and have been extended to swirling flew cases. A new 'AREA-PLOT' method, which unifies presentation of performance information, for plane and conical diffusers has been proposed. An added attraction of this method is that it displays all three geometric variables of the diffusers. As swirl modifies the boundary layer it was necessary to have same knowledge of the growth of the boundary layer in the axial flew situation. This was achieved by extending the 'ROSS-FRASER' model as a closed form solution requiring only the initial boundary conditions. The predictions compare very well with published experimental results. The swirling flow case has been considered both mathematically and experimentally, the latter being studied through flew visualisation and measurement. An extensive survey of available literature, on theoretical and experimental work, has been presented with particular emphasis on areas not covered by previous surveys. The mathematical analysis was aimed at identifying the dominant parameters. The solution indicates the preferred coordinate system and the possibility of further extension. It has been shown that it is possible to represent the tangential velocity distribution in the diffuser by a family of exponential curves. Further analysis indicated that the divergence of the solid-body rotation core was parallel to the wall of the diffuser. Few visualisation studies have identified breakdown and non-breakdown areas in turbulent swirling pipe flow. The development of the various modes of breakdown have been recorded. Detailed flew measurement in the 10° diffuser indicates that swirl has a definite effect on eliminating separation tendencies. It was found that swirl modifies the wall static pressure drop in the inlet pipe immediately upstream of the diffuser.