Destructive vibrations occur in industrial sieve-tray columns under certain flow conditions. The vibrations result from regular pulsations of the gas flow and pressure in the column. These pulsations have been studied with a 3 m. high perspex model sieve-tray column having a single active test-tray using air and water as process fluids. The effects on the amplitude and frequency of the pressure pulsations of gas velocity, tray liquid head, liquid crossflow and tray and column geometry were determined using pressure transducers with real time signal analysis. Pulsations were produced with fifteen tray geometries, including five hole diameters between 4.76 mm and 15.87 Mm. Most of the pulsations were produced at gas velocities below 12 mls and liquid heads below 200 N/m2. The pulsation fundamental frequency varied between 12 Hz. and 40 Hz., with an r.m.s. amplitude of up to 60 N/m2• The gas-liquid behaviour on the sieve-tray was studied using high speed cine-photography and electrical conductivity probes. This showed conclusively that, during the occurrence of pulsations, the gas-liquid interaction was highly regular and synchronised with the pressure pulsations. Detailed measurements were obtained of the degree of synchronisation and of the liquid motion on the tray. Several modes of gas-liquid interaction were identified. During pulsation production, the pulsating jet and the imperfect bubble were most common. Calculated fluctuations in the gas flow rate through the test-tray based on measured pressures agreed well with the results of the film analysis. A model of the synchronisation process is proposed based upon control by the pulsating jet. The model explains the measured limits of pulsation occurrence and permits some prediction of these limits for systems other than air-water. A simplified system flow analysis accounts for effects of gas density and column geometry on the pulsation frequency. The form of the empirical correlation of frequency with experimental variables is consistent with a physical description of the pulsating jet.

The thesis presents an intellectual history of air. It investigates a critical period when the concept of air changed from being an all pervasive 'element' within a predominantly Aristotelian cosmology, to an 'ocean', or a fluid and particulate body with mass and weight. The thesis is set in the context of the seventeenth-century revolution in science in England and is focused on the pneumatic work of Robert Boyle. The question behind the thesis is raised by a specific experiment published by Boyle in 1672 when he describes how he tried, and failed, to produce gnats in a flask that had been evacuated of air by his air-pump. The historical aspect of the thesis examines the development of the vacuum, a new and revolutionary experimental site, in tandem with the equally revolutionary developments in physiology. The philosophical aspect of the thesis examines the conceptual ideas being played out in the Gnat Experiment and the relation between natural philosophy and theology. In terms of its empirical method the experiment was emblematic of the new science being developed by Boyle. The ambition behind the experiment, however, and Boyle's disappointment at its failure, engages another level of enquiry. Of particular interest is the problem of 'thinking matter' and the conflicts it provoked in relation to discussions of air and the vacuum, life and soul. Though reignited by Descartes, the discussion can be traced back to the early theories of air in Presocratic philosophy and the development of the 'pneumatic tradition' through later Socratic and Stoic philosophy, as well as Christian theology, in the guise of pneuma. It becomes apparent that Boyle's 'air' engages a complex field of concepts and arguments that can be traced back to the beginnings of philosophy and science, and that are still burning.

No description available.

No description available.

No description available.

A novel program for the solution of one dimensional unsteady gas flow by the field method of characteristics is developed. It is shown how this was used as a design tool to make a simple pulsing combustor with associated fuel system. The results from running the actual combustor are used to refine assumptions used in the program. It is shown that qualitative agreement can be achieved between predicted and measure pressure waveforms. A design of collector/smoother is suggested and this design is tested and the results compared with a computer prediction. Again good waveform agreement is achieved. The experimental and computed results are used to suggest design improvement.

This investigation is concerned with the determination of high temperature gas reaction kinetics (primarily of acetylene decomposition) by using flame parameters. Steady-state flame propagation theory is employed to deduce overall activation energies and orders from the variation of burning velocity with temperature and pressure respectively. The development of experimental technique and apparatus is presented which made possible, for the first time, the stabilisation of an acetylene decomposition flame on a burner. To allow operation under these somewhat hazardous conditions, specially developed equipment is described which includes a safe acetylene compressor, sintered plate flow meters, and an annular burner. The theory and practice of this method for deducing overall kinetics is tested first on premixed propane- and ethylene-air flames. Several flame propagation theories are adapted in an attempt to obtain useful expressions for determining kinetics from measured variations in burning velocity. The theory of Semenoff, Fratk-Kamenetsky and Zeldovich is shown to be the most suitable and capable of predicting overall kinetics for hydrocarbon-air combustion, which are in good agreement with values reported in other investigations. It is also shown that by considering a specified intermediate reaction rather than the complete reaction, overall flame propagation theory may be capable of deducing actual radical kinetics. From observations and results obtained from acetylene, the reaction processes taking place in a freely propagating acetylene decomposition flame are shown to be distinctly different from those occurring in low temperature pyrolysis or shock waves. It is proposed that this anomaly is caused by dissociated hydrogen molecules and their subsequent diffusion and reaction. Finally, a reaction mechanism is postulated which accounts for the results.

No description available.

The work described in this thesis was aimed at obtaining and interpreting a data set for fully developed upwards air-water annular flows. The experiments were carried out in a vertical 32 mm diameter tube which was about 600 diameters long. The experiments were conducted at three fixed outlet pressures of 2.4,3.6 and 3.8 bars, respectively. The ranges of dimensionless gas and liquid flowrates covered were: 0< V/G < 10; 0< V/G <10; 0<V/L < 1.8. These flowrates considerably extended the range of previous air water measurements, which were generally confined to V/G < 2.5 and V/L < 0.3, and have allowed the Roberts and Hewitt (1969) flow pattern map to be considerably modified. Pressure gradient measurements were made using a purged D. P. cell system while film flowrate measurements were made by extracting the film through a porous wall. Wall shear stress measurements using hot film probes were also made and a method of calibrating the probes using a two-phase annular flow was proposed. The film flowrate results (Figures 3.1 to 3.14) indicate that there is a limiting film flowrate below which no significant entrainment occurs. The most interesting of the experimental results obtained, however, was the existence of a maximum in the plot of pressure gradient versus gas flowrate. This maximum occurred at high liquid and gas flows and was distinct from previously reported maxima and minima associated with the slug-churn-annular flow transitions, as shown in Figure 4.6. It was found that the newly discovered maximum was identical to the pressure gradient phenomenon in steam-water flows called the hydraulic resistance crisis. The obtained film flowrate and pressure gradient data were compared with a number of empirical correlations. None of the tested correlations could predict the data to any extent. Relative deviations between the predicted results and the experimental data were as great as 1900% for film flowrate and 200% for pressure gradient. A proposed entrainment correlation predicts well air-water data obtained in tubes ranging in diameter from 6 mm to 125 mm. A new annular flow model is also proposed which attempts to account for the effect of droplet turbulence suppression as well as the effect of waves on the interfacial roughness.

Flow control in the modern meaning of the term, is a technology that enables gains in performance, greater than that achievable using conventional design tools. To maximise the potential of this technology, forms of actuation have to be developed that require low power, do not modify the structure of the vehicle considerably, are reliable and above all efficient and effective in controlling the flowfield. A promising form of actuation is known as the synthetic or massless jet, so called because for an axi-symmetric orifice, a jet is formed from a train of vortex rings, with zero net mass flux. The synthetic jet actuator often consists of an orifice plate from which the vortex rings are formed, mounted to a cavity with a diaphragm at the other end. The periodic oscillation of the diaphragm causes the roll-up of a vortex ring at the orifice exit, that for sufficient levels of forcing, convects away under its self-induced velocity, before the next suction stroke commences. The present research is focused upon the fundamental understanding of the structure of an axi-symmetric synthetic jet embedded in a turbulent boundary layer, with the aim of understanding how non-dimensional parameters of practical importance affect the structure and the dynamics of the synthetic jet formation. A series of basic experiments that gradually added the salient features of an embedded synthetic jet were undertaken, and demonstrated that parameters such as the jet velocity ratio and Strouhal number have a large effect upon the dynamics and structure of a synthetic jet subjected to quiescent conditions, a cross-flow and shear in the form of a turbulent boundary layer. A model of a synthetic jet embedded in a boundary layer has been proposed based upon the results of these experiments, and hypothesises that the ejected vortex rings form two periodic counter-rotating streamwise vortical structures, provided that the vortex rings do not penetrate beyond the boundary layer. The effectiveness of synthetic jets in delaying the separation of a turbulent boundary layer from a circular cylinder by up to 5% in azimuth, was demonstrated. Surface flow visualisation provided further evidence of the presence of two periodic counter-rotating streamwise structures, and that three-dimensionality of the boundary layer near to separation, has a large effect on the structure of the synthetic jet and its interaction with the separation.