| dc.description.abstract | 1. Introduction
2. Experiments
2D PIV measurements were carried out in the near flow field of 2:1 (major axis: minor axis) aspect ratio elliptic jets issuing from both nozzle and orifice. The nominal jet exit velocity of 22 m/s provided a Reynolds number of 72,000 based on the major diameter.
PIV recordings were performed in both the major and minor planes spanning eight major diameters of the nozzle. A dual-cavity high-power Nd:YAG laser illuminated the measurement plane, and digitization of the particle image was carried out using a Kodak 1K × 1K cross-correlation CCD camera. A Pentium PHII 500 MHz PC interfaced to an IDT® (Integrated Design Tools Inc., USA) controller was used for synchronization of laser and image acquisition.
Cross-correlation using an interrogation cell size of 24 pixels resulted in 3750 2D vectors per image pair. For validation of PIV results at higher velocity moments, recordings were also carried out by varying the laser pulse interval and imaging area. A comprehensive semi-automated post-processing program developed in MATLAB was used for the analysis of time-averaged turbulence and instantaneous flow fields.
3. Results and Discussions
3.1 Mean Flow and Axis-Switching
The first part of the study documents the mean near-flow field development in elliptic nozzle and orifice jets. All mean flow profiles in the major and minor planes were obtained with the standard interrogation cell size of 24 × 24 pixels, providing a resolution of about 4.5 mm.
Comparison of mean streamwise velocity (U) profiles with those obtained by hot-wire anemometry in the nozzle showed excellent agreement both at the exit and across the shear layer beyond the potential core. Scatter in the lateral velocity (V) and transverse velocity (W) was low and followed expected trends, despite these components being an order of magnitude lower than the streamwise velocity.
The low scatter and higher accuracy in mean values were largely due to proper illumination, optimal seeding, and a high dynamic velocity range of over 300 (in the chosen magnification and interrogation cell size).
The mean streamwise velocity decay along the centerline showed a slightly shorter potential core and faster decay for the orifice.
The mean jet growth rate in the contoured nozzle showed a linear increase in the minor plane and nearly constant width in the major plane.
The orifice jet exhibited steeper growth in the minor plane and initial negative growth in the major plane.
The first axis-switching occurred at about 6.2 and 2.5 semi-major diameters in the nozzle and orifice jets, respectively.
3.2 Turbulence and Sensitivity of Parameters
This part of the study focuses on accurate estimation of turbulence intensity in the near field, considering spatial resolution, ensemble size, and flow gradient. A detailed sensitivity analysis was carried out in nozzle flow to resolve discrepancies in turbulence intensity and Reynolds stresses observed near the jet exit, where large flow gradients and wide ranges of turbulence intensities occur.
Velocity histograms were single-modal Gaussian type in all zones.
Convergence tests showed that a sample size of ~800 was adequate to provide streamwise turbulence intensity at 95% confidence level.
A sample size of over 1000 ensured good statistical stability.
Further analysis investigated the effect of local flow gradient and interrogation cell size (interdependent factors) on streamwise turbulence intensity. Large flow gradients within the interrogation cell significantly affected accurate estimation of turbulence quantities in the jet exit zone.
Accuracy of turbulence data improved with a proposed two-pass PIV processing scheme, which uses adaptive interrogation cell size in the second pass. The local mean displacement gradient across the cell obtained from the first pass was used to estimate cell size in the second pass. Threshold limits of gradient for different cell sizes were quantified.
This adaptive grid approach showed excellent comparison of turbulence intensities and Reynolds stresses in the high-gradient zone near the jet exit with hot-wire data.
Time-mean turbulence intensities were higher for the orifice in both streamwise and transverse directions compared to the nozzle.
Turbulence profile shapes were similar along the centerline in both nozzle and orifice, though axis-switching locations differed significantly.
3.3 Velocity Structures
In the third part of this work, 2-point spatial correlation and 2D filtering schemes were applied to study azimuthal velocity structures in nozzle and orifice jets.
2-point spatial correlation in the shear layer of the fluctuating streamwise component revealed structures of the order of integral length scales.
A novel 2D spatial filtering scheme (recently introduced in literature) was applied to extract velocity and vortical structures.
A 2D Gaussian kernel of appropriate size was convolved with the instantaneous flow field to obtain the spatially filtered velocity field.
The spatial filter was more effective in resolving large-scale structures compared to conventional time-averaged filtering.
Identification of these structures helps in understanding differences in jet development from nozzles and orifices, particularly the axis-switching phenomenon. | |
