No description available.

A fundamental study of two turbulent directly opposed impinging jets in a stagnant ambient fluid, unconfined or uninfluenced by walls is presented. By experimental investigation and numerical modeling, the main characteristics of direct impingement of two turbulent axisymmetric round jets under seven different geometrical and flow rate configurations (L*= L/d = { 5, 10, 20 }, where L is nozzle to nozzle separation distance and d is nozzle diameter, and Re = { 1500, 4500, 7500, 11000 }) are discussed. Flow visualization and velocity measurements performed using various laser based techniques have revealed the effects of Reynolds number, Re, and nozzle to nozzle separation, L, on the complex flow structure. Although locally valid, the classical analysis of turbulence is found unable to provide reliable results within the highly unstable and unsteady impingement region. When used to simulate the present flow, the assessment of the performance of three distinct k - epsilon turbulence models showed little disagreement between computed and experimental mean velocities and poor predictions as far as turbulence parameters are concerned.

Cette étude porte sur le développement d’une méthodologie de calcul pour évaluer les effets d’installation. Ce phénomène, qui représente le bruit d’interaction entre un jet double flux et un profil, constitue un problème modèle pour l’étude du bruit de jet installé en aéronautique. L’écoulement moyen est déterminé à partir de la résolution des équations de Navier-Stokes moyennées et du modèle de turbulence k - ω BSL de Menter. Lorsque le jet est isolé, il est possible ensuite d’utiliser le modèle de Tam & Auriault. Dans des configurations plus complexes, comme l’interaction jet-voilure considérée ici, la formulation statistique des termes sources de ce modèle est retenue à la place du modèle complet. Un propagateur acoustique, basé sur les équations d’Euler linéarisées, est utilisé pour compléter la modélisation. Dans un premier temps, une tuyère double flux avec plug est simulée et les résultats obtenus sont comparés aux données d’essais dans le but de valider la simulation numérique. De bons résultats sont obtenus. La même étude est ensuite réalisée sur une configuration installée, prenant en compte la même tuyère installée sous un profil. La simplicité de la configuration se justifie par le fait qu’il s’agit de développer une méthodologie de calcul permettant d’avoir un effet qualitatif de l’installation sur le développement du jet. En comparant les résultats aérodynamiques avec ceux obtenus numériquement pour la tuyère isolée, l’impact du profil sur le jet est mis en évidence au travers de la déviation du jet vers le profil et d’un déficit de l’énergie cinétique turbulente. La dernière étape consiste à caractériser l’impact de ces modifications sur les sources de bruit. Pour cela, le logiciel de propagation industriel Actran DGM est utilisé. Les sources calculées statistiquement par le modèle de Tam & Auriault sont introduites dans les équations d’Euler linéarisées. En première approche pour mettre en œuvre la méthodologie, les sources de bruit de jet sont assimilées à une distribution de monopoles équivalents. Cette modélisation permet de mettre en évidence les effets de diffraction et de masquage de l’onde causés par le profil. Le rayonnement en champ lointain est quant à lui obtenu avec la méthode intégrale de Ffowcs-Williams & Hawkings. / Excess noise induced by installation effects is numerically investigated in this work. Interaction noise between a turbofan jet engine and an airfoil is a simplified but relevant problem to address installed jet noise in aeronautics. The mean flow is determined from Reynolds-Averaged Navier-Stokes equations, using Menter k - ω BSL turbulence model. With jet only, fine scale turbulence model of Tam & Auriault can be used directly for jet noise prediction. To assess jet-wing interaction in industrial configuration, the statistical formulation of the source terms is retained from this stochastic model, and the acoustic propagation is performed using linearised Euler equations. A dual stream jet is firstly computed and the results are compared to available data in order to validate the numerical simulation. Satisfactory results are obtained. The same study is then performed on an installed configuration, taking into account the same nozzle installed under a profile. This first configuration is is used to develop a methodology of calculation by having a qualitative effect of the installation on the jet behaviour. By comparing the aerodynamic results with those obtained numerically for the isolated nozzle, the impact of the jet profile is highlighted through the jet deflection and a modification of the turbulent kinetic energy field. The last step is to characterize the impact of these modifications on the noise sources. For this, the industrial propagation software Actran DGM is used. The statistical formulation of the source calculated by Tam & Auriault model is introduced into the linearised Euler equations. As a first step, the jet sources are defined as a distribution of equivalent monopoles. This modeling allows highlighting the effects of diffraction and the masking effects caused by the profile. The radiation in the far field is obtained with the integral method of Ffowcs-Williams & Hawkings.

The effects of bubbles on a jet issued below and parallel to a free surface are experimentally studied. The jet under study is isothermal and in fresh water, with air injectors that allow variation of the inlet air volume fraction for 0% to 13%. Measurements of the jet exit conditions, water velocity, water entrainment, Reynolds stresses and surface currents have been performed using LDV, PIV and surface PIV. Air volume fraction, bubble velocity, chord length and free surface elevation and RMS have been obtained using local phase detection probes. Visualization was performed using laser-induced fluorescence. Measurements show that water entrainment decreases up to 22% with the presence of bubbles, but surface current strength increases up to 60% with 0.4 l/min of air injection. The mean free surface elevation and turbulent fluctuation significantly increase with the injection of air. The water normal Reynolds stresses are damped by the presence of bubbles in the bulk of the liquid, but very close to the free surface the effect is reversed and the normal Reynolds stresses increase slightly for the bubbly flow. Flow visualizations show that the two-phase jet is lifted with the presence of bubbles and attaches to the free surface sooner. Significant bubble coalescence is observed, leading to an increase of 20% in mean bubble size as the jet develops. The coalescence near the free surface is particularly strong, due to the time it takes the bubbles to pierce the free surface, resulting in a considerable increase in the local air volume fraction.

A fundamental study of two turbulent directly opposed impinging jets in a stagnant ambient fluid, unconfined or uninfluenced by walls is presented. By experimental investigation and numerical modeling, the main characteristics of direct impingement of two turbulent axisymmetric round jets under seven different geometrical and flow rate configurations (L*= L/d = { 5, 10, 20 }, where L is nozzle to nozzle separation distance and d is nozzle diameter, and Re = { 1500, 4500, 7500, 11000 }) are discussed. Flow visualization and velocity measurements performed using various laser based techniques have revealed the effects of Reynolds number, Re, and nozzle to nozzle separation, L, on the complex flow structure. Although locally valid, the classical analysis of turbulence is found unable to provide reliable results within the highly unstable and unsteady impingement region. When used to simulate the present flow, the assessment of the performance of three distinct k - epsilon turbulence models showed little disagreement between computed and experimental mean velocities and poor predictions as far as turbulence parameters are concerned.

In order to extend fluid-based flow control techniques that have been demonstrated at low subsonic speeds to high speed flows, it is necessary to develop actuators having sufficient momentum to control and manipulate high speed flows. Two fluidic actuation approaches are developed where the control jet may reach supersonic velocities and their performance is characterized. The first actuator is a compressible synthetic (zero net mass flux) jet. This is an extension of previous work on synthetic jets with an increase in driver power yielding substantial pressurization of the cavity such that the flow is compressible. The jet is generated using a piston/cylinder actuator, and the effects of variation of the orifice diameter, actuation frequency, and compression ratio are investigated. Operation in the compressible regime uniquely affects the time-dependent cylinder pressure in that the duty cycle of the system shifts such that the suction phase is longer than the blowing phase. The structure of the jet in the near-field is documented using particle image velocimetry and Schlieren flow visualization. In the range investigated, the stroke length is sufficiently long that the jet flow is dominated by a starting jet rather than a starting vortex (which is typical of low-speed synthetic jets). A simple, quasi-static numerical model of the cylinder pressure is developed and is in generally good agreement with the experimental results. This model is used to assess system parameters which could not be measured directly (e.g., the dynamic gas temperature and mass within the cylinder) and for predictions of the actuator performance beyond the current experimental range. Finally, an experiment is described with self-actuated valves mounted into the cylinder head which effectively icrease the orifice area in suction and overcome some of the limitations inherent to compressible operation. The second actuation concept is the combustion-driven jet actuator. This device consists of a small-scale (nominally 1 cc) combustion chamber which is filled with premixed fuel and oxidizer. The mixture is ignited using an integrated spark gap, creating a momentary high pressure burst within the combustor that drives a high-speed jet from an exhaust orifice. At these scales, the entire combustion process is complete within several milliseconds and the cycle resumes when fresh fuel/oxidizer is fed into the chamber and displaces the remaining combustion products. The actuator performance is characterized by using dynamaic measurements of the combustor pressure along with Schlieren flow visualization, limited dynamic thrust measurements, and flame photography. The effects of variation in the following system parameters are investigated: fuel type and mixture ratio, exhaust orifice diameter, chamber aspect ratio, chamber volume, fuel/air flow rate, ignition/combustion frequency, and spark ignition energy. The resulting performance trends are documented and the basis for each discussed. Finally, a proof-of-concept experiment demonstrates the utility of teh combustion-driven jet actuators at low-speed for transitory reattachment of a separated flow over an airfoil at high angles of attack.

An experimental investigation was undertaken to test the effectiveness of a novel design for controlling the aerodynamics of an airfoil. A synthetic jet actuator (SJA) was placed inside a NACA 0015 airfoil with its jet at 12.5% of the chord length, hereby referred to as the leading edge actuator. Four centrifugal fans across the span were mounted at 70% of the chord and the jet formed by them was located at 99% of the chord, hereby referred to as the trailing edge actuator. The effects of these actuators on the aerodynamic properties were studied, separately and then in conjunction, with varying angles of attack. The leading edge actuator delays the onset of stall up to 24 degrees, the maximum angle of attack that could be attained. The control of the aerodynamics was achieved by controlling the amount of separated region. There was no effect of the actuation at lower angles of attack. The trailing edge actuator provides aerodynamic control at both low and high angles of attack. The study investigated the effect of jet momentum coefficient on the aerodynamic properties for various angles of attack. The data obtained shows that lift control (in both positive and negative direction) was achieved even at low angles. The actuator enhances the aerodynamic properties by changing the pressure distribution as well as by delaying flow separation. Study of the combined actuation shows that the synthetic jet actuator was very effective in delaying stall when the trailing edge jet was ejected from the upper surface. For the case when the jet is ejected from the lower surface, there is less control. This can be accounted for by the difference in aerodynamic loading for both cases.

Flameless combustion in a 48 kW pilot scale furnace fired with natural gas is studied experimentally and computationally. The burner geometry involved a tunnel furnace with two separate feed streams --- one for a high momentum air jet and the other for a low momentum fuel jet. This burner configuration, called a Strong-Jet/Weak-Jet (SJWJ) burner, together with the jetto- jet interactions generate the flameless combustion mode with relatively uniform furnace gas temperature distributions and low NOX emissions. Experiments were carried out under laboratory conditions for turbulent reactive mixing in order to obtain local temperature and gas concentrations. The experimental findings were used to test the performance of CFD numerical models for turbulence, mixing and chemical reactions. For the SJWJ furnace operated in flameless combustion mode, 32 different flow cases were examined to assess the effects of the three main parameters (fuel/air momentum flux ratio, fuel/air nozzle separation distance and fuel injection angle) on the furnace wall temperature profile. Three specific flow configurations were selected for detailed near-field temperature measurements. The gas temperature distribution inside the combustion chamber was found to be relatively uniform, a characteristic of flameless combustion. Four flow configurations were studied to examine the effect of the fuel jet injection angle (0 degrees or 10 degrees) and fuel/air momentum flux ratio (0.0300 and 0.0426) on the mixing, combustion performance and NOX emissions. Gas compositions were measured in the flue gas and within the furnace at selected locations to estimate the concentrations of CO2 CO, CH4, O2, NO and NOX. The NOX concentrations in the flue gas were quite low, ranging from 7 - 13 ppm, another characteristic of flameless combustion. The combusting flow CFD calculations were carried out using the k-ε turbulence model and the eddy-dissipation model for methane-air-2-step reactions to predict the temperature and concentration field. The numerical results for gas temperature and compositions of CH4, O2 and CO2 generally showed good agreement with the experimental data. The predicted CO concentration profiles followed expected trends but the experimental data were generally underpredicted. The NOX concentrations were estimated through post-processing and these results were significantly underpredicted. / Thesis (Ph.D, Mechanical and Materials Engineering) -- Queen's University, 2008-04-04 11:25:25.455

Various sensing and digital signal processing approaches to detect cavitation in a water jet propulsion unit were examined based on results in the literature. Several commercially viable sensors were evaluated based upon their ability to detect the cavitation phenomenon, cost, and robustness. An algorithm has been implemented and tested against data recorded from the candidate sensors. The combination of vibration and pressure sensors and the algorithm appear promising and a path for further development and testing is available to Hamilton Jet.

Jet in crossflow (JICF) configurations are often used for secondary fuel injection in staged-fuel combustion systems. The high temperature, vitiated crossflow in these systems is inherently unsteady and characterized by random, turbulent fluctuations and coherent, acoustic oscillations. This thesis presents the results of an experimental investigation into the dynamics of non-reacting and reacting jets injected into unsteady, vitiated crossflow. The flow structure and flame stabilization of jets with different momentum flux and density ratios relative to the crossflow are characterized using simultaneous time-resolved stereoscopic particle image velocimetry (SPIV) synchronized with OH planar laser induced fluorescence (PLIF). A modified trajectory scaling law is developed to account for the influence of near-field heat release on the jet trajectory. The second part of this work focuses on the response of a JICF to crossflow forcing. Acoustic drivers are used to excite natural resonances of the facility, which are characterized using the two-microphone method. Spectral analysis of SPIV results shows that, while the jet response to crossflow velocity fluctuations is often negligible, the fluctuating crossflow pressure induces a significant fluctuating jet exit velocity, which leads to periodic jet flapping. The flame response to crossflow forcing is studied using flame edge tracking. An analytical model is developed that predicts the dependence of the jet injector impedance upon important JICF parameters. In the final part of this work, vortex tracking and Mie scattering flow visualization are used to investigate the effect of near-field heat release on the shear layer dynamics. A phenomenological model is developed to explain the effect of combustion on the shear layer stability of density stratified, reacting JICF. The results of this study demonstrate the important effects of near-field heat release and crossflow acoustics on the dynamics of reacting JICF.