The laser diffraction technique constitute an easy and fast on line measurement of drop sizes. Based on the well known theory of diffraction, a laser diagnostic technique is described for the measurement of droplet size distribution. Limitations of the technique are studied to identify accuracy of measurements when applied to dense sprays (i.e. diesel fuel). Drop size measurements are then conducted for different diesel fuel under different conditions of injection and chamber pressures. Results lead to better understanding of the atomization process.

The aim of this experiment was to analyze the interaction and the development of two flows in the rectangular confined cavity described in this paper. The first was a forced convective flow generated by injecting air at low temperatures inside the cavity in a horizontal direction through an opened area located in the cavity front. Inside this cavity, a natural convective plume was developed simultaneously. Different experimental techniques were used in order to describe either the mean and fluctuating air flow or the mean and fluctuating temperature field, and also to analyze the interaction between these two flows. Firstly, our set-up was validated with the free jet configuration: we obtained a well-known three-dimensional turbulent free jet. Without plume, analysis of the results showed that injection flow behavior corresponded to a characteristic 3D wall jet along the upper wall, with an 15400 injection Reynolds number. The Coanda effect maintained the jet along the upper wall. We determined the parameters of the main velocity law in relation to those detailed in the literature. After that, the main flow developed simultaneously in the longitudinal and vertical directions, so that a recirculation zone appeared. With the plume that developed over an obstacle, we observed that cold flow behavior changed, because of the combination of some of the main parameters: geometrical parameters, injection Reynolds number and temperature gradient between injected flow and plume. For example, with an injection Reynolds number equal to 15400, we measured the influence of the plume on the main flow through the changes in the dynamic and temperature profiles. Three regimes were obtained: the main flow was maintained and no natural convective plume was observed; the transitional regime was characterized by the main flow development along the obstacles, in spite of the development of the natural convective plumes; the natural convective flow modified the wall jet and unsteady classical natural convective “mushrooms” were observed. The topical results deal with the parametric study. We worked on a critical number characterizing the mixture of the jet and the plume. This number is defined as the ratio of the Grashof number to the Reynolds number of the mixture flow. When this critical value was exceeded, the natural convective plume was able to stop the inlet wall jet.

This investigation examines the surface-pressure fluctuations and associated flow structures spatiotemporally in the developing flow downstream of the reattachment point of a fence-with-splitter-plate flow. Simultaneous measurements of the wall-pressure and velocity field were undertaken using a 16-microphone array, extending over the streamwise range 1.67 < x/X r < 3.33 (where X r is the mean reattachment length), and X-hotwire sensor at two Reynolds numbers of 8000 and 16000, based on the fence height above the splitter plate. The array data were used to obtain the wavenumber-frequency spectrum of the wall-pressure fluctuation. The results illustrate that Taylor hypothesis of frozen eddies reasonably describes the flow in the investigated zone. This allowed utilization of the time-dependent LSE of the velocity field to estimate the spatial structure of the flow above the microphone array. The results confirm the association of the most-energetic pressure fluctuation with the passage of quasi-periodic vortices.

Noise generated from large-scale wave-like eddies in a Mach 0.9 hot axisymmetric jet is studied. The mean jet flow is computed using a Reynolds-Averaged Navier-Stokes solver with the k-ω turbulence model. Spatial development of near-field pressure perturbations is computed using a 3D Parabolized Stability Equation (PSE) method, and the far-field noise radiated from these convective instabilities is obtained by solving the wave equation. The 3D PSE method developed here allows the effects of strong azimuthal mean-flow variations to be captured and analyzed. Results show that large-scale wave-like eddies, or instability waves, travel slightly above sonic speed for the first few jet diameters, and that the dominant noise sources are concentrated near the edge of the time-averaged potential core. Good agreement between computed results and experiment are found for relative sound pressure levels and directivity. Also, at the low frequencies considered, the sound field associated with azimuthal wave-number m = 0 shows better agreement with experimental data than with (the most amplified) m = 1.

Recent studies on automotive engine cooling fan systems carried out in partnership between LEMFI and Valeo have led to the definition of a range of efficient stator designs. It was shown that an adequate rotor-stator coupling could yield significant efficiency gains over the whole range of Valeo fans with diameters ranging from 280 mm to 460 mm. Efficiency gains ranging from 12 percentage points for the 350 mm fan diameter to 2 points for the 460 mm were estimated by using a simplified radial equilibrium design (SRE). These predictions have been verified experimentally by recent designs for various fan system diameters. The present study then describes a newly developed aeroacoustic test facility dedicated to automotive fan systems. The influence of the rotor-stator coupling is then shown by comparing the rotor alone and rotor-stator configurations, not only on the overall performances and velocity fields at the system outlet measured by a 5-hole probe, but also on the noise generated by the fan systems for various flow conditions. At the nominal flow rate, the efficient stators are shown to bring little or no extra noise if the number of stators and the rotor-distance are carefully chosen. At other flow rates, additional noise might be expected.

A spectrally resolved molecular Rayleigh scattering technique was used to simultaneously measure axial component of velocity U, static temperature T, and density ρ in unheated free jets at Mach numbers M j = 0.6, 0.95, 1.4 and 1.8. The latter two conditions were achieved using contoured convergent-divergent nozzles. A narrow line-width continuous wave laser was passed through the jet plumes and molecular scattered light from a small region on the beam was collected and analyzed using a Fabry-Perot interferometer. The optical spectrum analysis provided measures of velocity and static temperature. The local air density at the probe volume was determined by monitoring the intensity variation of the scattered light using photomultiplier tubes. The Fabry-Perot interferometer was operated in the imaging mode, whereby the fringe formed at the image plane was captured by a cooled CCD camera. Special attention was given to remove dust particles from the plume and to provide adequate vibration isolation to the optical components. The velocity profiles from various operating conditions were compared with that measured by a Pitot tube. An excellent comparison within 5m/s demonstrated the maturity of the technique. Temperature was measured least accurately, within 10K, while density was measured within 1% uncertainty. The survey data consisted of centerline variations and radial profiles of time-averaged U, T and ρ. The static temperature and density values were used to determine static pressure variations inside the jet. The data provided a comparative study of jet growth rates with increasing Mach number. The current work is part of a data-base development project for Computational Fluid Dynamics and Aeroacoustics codes that endeavor to predict noise characteristics of high speed jets. A limited amount of far field noise spectra from the same jets are also presented. Finally, a direct experimental validation was obtained for the Crocco-Busemann equation which is commonly used to predict temperature and density profiles from known velocity profiles. Data presented in this paper are available in ASCII format upon request.

This paper studies the effect of two given free-stream turbulence intensities on film cooling at different blowing ratios on one cascade, made up of five blades. Experiments on the suction side were done at three different blowing ratios 0.5,1.0 and 1.5; those on the pressure side were done at four different blowing ratios, 0.5,0.75,1.5 and 1.75. The two given free stream turbulence intensities are 0.5% and 5% respectively. In these experiments, we adopted infrared thermography to capture the images of temperature distributions on blade surface. According to the experimental results, the turbulence acts on the effectiveness distributions differently on the pressure side and suction side due to the existence of blade curvature. Bigger turbulence intensity can worsen the cooling effect on both the suction side and the pressure side.

In this investigation an intensive experimental study was conducted to study the flow quality in a test section of a low speed, V ∞ = 10–100, closed return wind tunnel. An interesting result was obtained when a trip strip, a guitar wire with D = 1.14mm, was installed near the end of the contraction region. Addition of this trip wire resulted in a turbulent reduction almost at all speed ranges along the entire test section. Further, it smoothness the variation of turbulent intensity along the test section too. Velocity profiles along the tunnel wall at various speeds using a rake were measured with and without the trip wire. In addition, pressure contours at different location of the test section was measured too.

SPLNoise is an aeroacoustic program developed to predict tonal noise generated by low speed axial fans. It is based on a generalization of the acoustic analogy of Lowson. The main input to the program are the radial thrust and torque distribution along the blade. This radial distribution is obtained from a CFD calculation. An automatic link to CFX commercial CFD code was made. Noise testing was performed in the anechoic and reverberant room at the Trane Company. Comparison between predicted noise levels and experimental data is very satisfactory.

We have developed a prediction model for the overall level of aeroacoustic noise generated by cross-flow fans. It is based on two assumptions: the velocities along the outside edge of the impeller have an especially close relation to the noise levels, and the noise levels are in proportion to the sixth power of the relative velocities at the edge of the blade. A computational fluid dynamics is used to obtain the necessary velocity-field data for this prediction model. The predicted noise levels of a test impeller for different flow coefficients are in good agreement with the measured results. This means that variations in noise levels with the flow coefficient can be described accurately by the prediction model, which should prove to be a useful tool for speeding up the development of silent cross-flow fans.

A commercial CFD program and boundary layer analysis are used to design the inlet of an open circuit wind tunnel for aerodynamic testing of full scale, class eight trucks. The goal of the study is to obtain the combination of wall shape, contraction length, and setting screens that provide a uniform velocity profile upstream of the vehicle under test. Thwaites method is used to verify that the boundary layer does not separate. Velocity field predictions upstream of the vehicle under test are presented for different design configurations of the inlet. A variation in the velocity magnitude of less than one percent outside of the boundary layer is achievable. The contraction introduces a weak secondary flow in the corners of the wind tunnel, but the cross stream components do not lead to significant distortion of the axial profile.

Reduction of aerodynamic drag of heavy-duty trucks can significantly save fuel costs and US dependence on the imported oil. Reduction of aerodynamic drag by 30% can result in fuel cost savings in billions of dollars every year. Aerodynamic drag of truck depends on the frontal cross-sectional area and the speed of the vehicle. In addition, the gap between the cabin and the trailer significantly affect the drag of the truck. This paper investigates how changes in the cabin geometry and the trailer gap can reduce the aerodynamic drag using numerical simulations. The numerical simulations were carried out using Computational Fluid Dynamics (CFD) software, CFX-5.5, from AEA Technologies (now owned by ANSYS). Effects of vehicle speed, cabin geometry, and trailer gap on the aerodynamic drag were investigated.