This thesis is a study of quantum turbulence in superfluid helium. Superfluid he- lium consists of two interpenetrating fluids, a viscous normal fluid and an inviscid superfluid, coupled by a mutual friction. The thesis is divided in two parts. The first part deals with fully developed turbulence. In analogy to classical turbulence theory, I develop a two-fluid shell model to study superfluid turbulence. I investigate the energy spectra and the balance of fluxes between the two fluids as a function of temperature in continuously forced turbulence and show how both fluids gener- ate the classical k−5/3 Kolmogorov scaling law within the inertial subrange, whilst simultaneously exhibiting deviations from this law outside the subrange due to the mutual friction force. I furthermore investigate the decay of turbulence in the ab- sence of forcing. I compare my results with experiments and existing calculations. I find that, at sufficiently low temperatures a build-up of energy develops at high wavenumbers suggesting the need for a further dissipative effect, such as the Kelvin wave cascade and phonon emission. The second part of this thesis is concerned with complex vortex flows. It is well known that two coaxial vortex rings can leapfrog about each other. By direct numerical simulation, I show that in superfluid helium the effect can be generalised to a large number of vortex rings, which form a toroidal bundle. The bundle is shown to be robust, travelling a significant distance compared to its diameter, whilst at the same time becoming linked and turbulent. I also discuss the effects of friction at non-zero temperatures, and show how in this case the presence of normal fluid rotation is necessary for the stability of the bundle. Although I am unable to model numerically the number of rings realised in experiments, I compare my results with those of experiments both qualitatively and by extending the equations for a single quantised vortex ring.

The phenomena of AC electroosmosis (ACEO) and dielectrophoresis (DEP) which are AC electrokinetics (ACEK) phenomenon but the earlier is always referred as AC electrohydrodynamic (AC ERD), have been used for many years to study, manipulate, trap, separate and collect particles of different scales and origins. The techniques have been adopted by builders of Microelectromechanical systems (MEMS) for Lab on a Chip (LaC) and Micro Total Analysis System (flTAS) for developing microfluidic devices. In this work, electrodes have been used to investigate these phenomena, to understand and optimise electroosmotic (EO) trapping. The effect of medium viscosity, medium density and temperature has been examined. It has been shown that medium viscosity is dominant over medium density as factor that affects EO trapping. It is also has ruled out centrifugal force as the dominant factor of EO trapping. Additionally, this work has investigated the effect of chamber height and electrode size on the efficiency of EO trapping.

The structure of turbulent flow in vertical upwards annular air water two-phase flow was examined. Experiments were carried out in a 32 mm internal diameter tube using laser Doppler anemometry. Simultaneous measurements of the two velocity components and the Reynolds stress were obtained by the use of two colours (blue and green) of a 50 mW argon ion laser. The gas core was seeded by polystyrene particles of 1 um diameter which were believed to follow the gas turbulent fluctuations. The characteristics of the signal were used to discriminate these tracer particles from the water droplets. The gas velocity profiles were shown to be more peaked at the centre of the tube than those observed in turbulent single phase flow. Comparative analysis with other data suggested that both interfacial roughness and, particularly, the momentum interchange between the droplets and the gas core, are the most important factors affecting the gas velocity profile in annular flow. Turbulent fluctuations of the gas velocity were found to be significantly higher than those typical of single phase flow, for similar gas Reynolds numbers. The interfacial shear, droplet size and concentration and the presence of disturbance waves at the interface were identified as being the most important factors affecting the gas turbulence in annular flow. A model was developed to predict the axial component of the turbulent fluctuations at the centre of the tube. The turbulence transport properties were observed to differ from those typical of single phase flow: i.e., higher production of turbulent energy (associated with higher anisotropy ratios), higher turbulence length scales and comparativelly lower dissipation ratios. Extrapolation of the mixing length theory to annular flow appeared to be inappropriate. Droplet size measurements showed that the gas velocity and the droplet concentration are the most important parameters affecting droplet size. At low droplet concentrations (where the gas-droplet interaction is more important than that between the droplets), a modified Weber number based on the homogeneous gas core momentum describes the maximum droplet diameter. At high droplet concentrations, the data suggests that coalescence is the dominant factor. Droplet velocity was found to be related to the size of the droplets: i.e., large droplets travel slower than small ones. The difference in velocity between large and small droplets was found to depend on the liquid and gas flow rates. This observation is related to conditions where droplet coalescence occurs. The effect of inserts on droplet size and the entrained fraction was examined. Disturbances in the channel geometry were found to affect the mean droplet size due to the creation of a new droplet population. The entrained fraction of liquid downstream of the insert was also affected. A model was formulated to describe the liquid interchange in the presence of a vertical plate.

No description available.

The basic principle of the cross-correlation method of flow measurement is described with diagrams and mathematics. The subject of the thesis is put into perspective by a brief description of previous work in the subjects of fluidynamics and correlation flowmetering. Attention is brought to previous work done in the particular field of ultrasonic correlation flowmetering. By referring to previous work in fluidynamics, signal handling and correlation flowmetering, the author shows that there are more factors influencing the ultrasonic correlation flowmeter's performance than have previously been considered. He develops a discussion about these factors and points out how they might be expected to influence the flowmeter's performance. The flowmeter's sensor head and the acoustic sensors are illustrated and described. There is a brief description of the general purpose, laboratory correlator and the associated spectrum display unit which was used for many of the experiments. The mode of action of the industrial, self tracking, two-point-difference, flowmeter correlator is outlined. A detailed section on the sensor electronics describes the reasons for choosing phase demodulation, how a phase demodulator works and the design and reasons for the novel sensor electronics devised by the author. The flow rig is described and illustrated. Pour preliminary experiments are described together with results which show their outcome. The experiments are; the determination of the sensor electronics phase/voltage transfer characteristics, observation of the sensor's resonances and acoustic standing waves in the metered fluid, demonstration that a novel feature of the author's sensor electronics was necessary and worked and demonstration that the flowmeter was capable of working on perfectly clean liquids.

No description available.

Coherent turbulent structures in the boundary layer resulting in turbulent bursts penetrating the sublayer, have been proposed as a mechanism for the removal of fouling deposits. The interaction between the bursts and representations of particulate and biological fouling deposits was studied using dye to highlight the flow structure and a twin laser system to provide illumination for filming and video recording. Over a smooth plate the dimensionless spanwise low speed streak spacing (λ+s = λsU*/v) was found to be 93 which is in excellent agreement with the literature. Grooved plates and polysaccharide gums were used as test surfaces and microglass beads and biofilms were also employed. The mean cleaning efficiency of bursts over a particulate deposit was found to be 0.023 percent and it was considered that a burst event had insufficient energy to remove fouling. Dimensionless streak spacings over the grooves ranged from 83.1 to 196.7 for longitudinal grooves and from 75.9 to 209.9 for transverse grooves. This increase in the spacing was expected from the literature. A change in the scaling parameters from inner variables over a smooth surface to outer variables over a grooved plate was recorded. Burst counts over the grooves, both longitudinal and transverse, and the gums were found to increase from the smooth surface. However, the spanwise spacing of low speed streaks also increased Over the grooves bursts were noted to display shallower ejection angles than over a smooth surface. A typical burst angle over longitudinal grooves was 3.70 compared to 4.30 for a smooth surface. This together with an increase in streak persistence and a delay in transition to turbulence suggested that the surfaces representing fouling deposits reduced the energy of burst events so making removal by burst action unlikely. Further work is recommended to investigate the flow over biofilms grown under flowing conditions, and over random roughnesses. The determination of the burst duration and intensity and the skin fiction over these surfaces would be a useful undertaking.

Environmental hydraulics covers a very wide range of applications including free surface flows in rivers. estuaries and lakes. To find engineering solutions to environmental hydraulics problems. 3D numerical modelling is nowadays widely used. However. the computation of sharp spatial gradients (such as found in stratified estuaries and lakes. around plumes near outfalls along rivers and coasts or in exchange areas of high shear). and the modelling of these processes along steep bathymetric slopes (such as found at the edge of dredged channels or of the continental shelf) remains a challenge. In addition. crude assumptions (such as the hydrostatic assumption) are often made to the primary differential equations in order to simplify the problem and enable long term prediction of environmental hydraulic changes. In this thesis. a robust adaptive mesh displacement (AMD) method is implemented and validated against the lock exchange case in particular. The AMD method aims at vertically focusing nodes within each water column to capture sharp gradients. while reducing the number of nodes or requiring prior knowledge of the flow structure. Second. a direct computation of dynamic pressure is introduced based on the equation of vertical momentum and validated against the analytical potential flow theory solution of a source-sink pair. Dynamic pressure is necessary to model destratification recirculation devices. or flow over dredge channel. or solitary waves. for instance. This direct computation method makes the hydrostatic assumption redundant. Third. a new advection scheme is implemented. whose main advantage is simplicity averaging over Riemann problems without solving them. while excessive numerical viscosity is compensated for by using high-resolution MUSCL type reconstruction. Recommendations are made in this thesis to extend the advection scheme developed herein for tracer advection to the non-linear shallow water equations. to the diffusion terms and to turbulence closure laws within the same finite element framework.

This thesis uses numerical, asymptotic and flow structural techniques to examine various aspects of rotor blade flows and ground effect. It explores two- and three-dimensional flows, generally concentrating upon regimes that have a degree of relevance to typical rotor blade flows. Chapter 2 considers, as a first step towards understanding a general rotor blade system in ground effect, a finite rotating disc near horizontal ground. More specifically, it concentrates on determining the layer shape beyond the disc rim that, due to the presence of the ground, cannot remain flat without violating a pressure condition across it. Chapter 3 examines the flow past many blades in ground effect using both a numerical approach and considering various limits of interest to illuminate some of the important features such as enhanced lift and sheltering effects. Chapter 4 then extends this by exploring the many blade limit, whereby the flow develops a periodic structure once sufficiently many blades have been passed. We then move on to three-dimensional configurations. Chapter 5 takes the previous work further by considering the interactive case that arises after a very large number of blades have been passed, generating a pressure-displacement interaction in the boundary layer. We examine the case of three-dimensional blades, considering the full triple deck problem and then the short blade limit, investigating the flow structure for this physically relevant case. Chapter 6 considers the flow past a three-dimensional hump on a blade of a rotor, examining the flow structure and solution and tentatively using this to propose a description of the flow past the trailing corner of a typical rotor blade. Finally Chapter 7 returns to ground effect, exploring the flow past a single, three-dimensional blade near the ground. It uses a compact difference technique to examine the flow solution for a particular blade shape and investigates the idea of change-over points, where the effective leading edge becomes a trailing edge switching the boundary conditions, these points being generally unknown in advance

No description available.