Bulletin of the American Physical Society
2005 58th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 20–22, 2005; Chicago, IL
Session LQ: Jet Stability III |
Hide Abstracts |
Chair: Peter Schmid, University of Washington Room: Hilton Chicago Stevens 2 |
Tuesday, November 22, 2005 8:00AM - 8:13AM |
LQ.00001: Stability of reacting gas jets Joseph W. Nichols, James J. Riley, Peter J. Schmid The stability of a viscous, reacting, variable-density jet is analyzed by means of linear stability analysis and direct numerical simulation (DNS). The spatial branches of the transformed linearized low Mach number equations are solved using a matrix method to obtain a complete spectrum of eigenmodes. The effect of reaction is isolated by comparing results from the reacting jet to those from a non-reacting jet with the same mean profiles (in which the density is lowered in the region where the flame would otherwise be). While reaction serves to stabilize the flow at most frequencies, it is also found to excite a low frequency numerical instability which may mask physically relevant results from DNS if the mixture fraction is not explicitly constrained. Furthermore, this instability mode takes on physical significance in the case of partially premixed jets. Results from the stability theory are compared to results of direct numerical simulation of the fully nonlinear problem. In order to understand the effect of streamwise development of the jet on its stability, linear analysis is applied to mean profiles measured from DNS at various axial locations. [Preview Abstract] |
Tuesday, November 22, 2005 8:13AM - 8:26AM |
LQ.00002: The counter-rotating core of a swirling turbulent jet Luca Facciolo, P. Henrik Alfredsson A swirling jet is generated by a fully developed, turbulent rotating pipe flow (6 m long, diameter 60 mm). The Reynolds number is 24000 and the swirl number (ratio between the velocity at the pipe wall and the bulk velocity in the pipe) 0.5. Due to the effect of the cross-stream Reynolds stress the flow in a rotating pipe does not reach a solid body rotation and instead the azimuthal velocity component lags behind it. The jet issued at the pipe end preserves, as shown by both experiments and numerical simulations, the azimuthal velocity imposed by the pipe flow in the central region a few diameters downstream. Moving further downstream the azimuthal component decays until, in the core of the jet, it becomes negative and the flow rotates in the opposite direction with respect to the rotation of the pipe. The counter-rotating core covers a region of approximately the pipe diameter and its amplitude represents a few percent of the velocity at the pipe wall but is clearly detected in experiments and simulations. Hot-wire and LDV data show the development of the jet flow field and confirm the counter- rotating core at a distance of approximately 6 diameters from the pipe outlet. Time and space resolved stereo PIV data are used to analyse the structures of the counter-rotating core in the cross flow plane, among other things with the use of POD. [Preview Abstract] |
Tuesday, November 22, 2005 8:26AM - 8:39AM |
LQ.00003: Effects of collar cross-sectional shape on self-excited collared jets K.S. Tan, T.H. New, H.M. Tsai Hot-wire measurement studies are performed to understand the velocity fields and turbulence statistics arising from self-excited collared jets. The effects of azimuthal variations in the collar step height are studied by using collars of different cross-sectional shapes and lengths. Results show that as the collar length increases, general centerline velocity decay rates decrease initially but rise rapidly to a maximum. At their respective optimum collar lengths, the square collar is found to incur a higher velocity decay rate with a significantly longer collar, as compared to the circular collar. Furthermore, the presence of a collar leads to turbulence suppression and the inhibition of vortex pairing in the near field downstream of the collar, with a circular collar effecting higher turbulence suppression. The presence of a collar is also found to accelerate transition to fully developed turbulence. A circular collar leads to an abrupt increase in self-excitation amplitude due to its constant step height whereas a square collar promotes a more gradual rise in excitation with a lower amplitude maximum. Spectra analysis indicates that, as compared to the circular collar, the square collar results in a smaller variation of the dominant frequencies within each excitation stage which also extends over a longer range of collar length values. The effects of a triangular collar will also be examined. [Preview Abstract] |
Tuesday, November 22, 2005 8:39AM - 8:52AM |
LQ.00004: Instability of jet plume from an overexpanded nozzle Dimitri Papamoschou, Paul Rossetti Our study involves the phenomenon of supersonic nozzle flow separation wherein a shock forms inside a convergent-divergent nozzle. Of particular interest is the instability of the jet plume exiting this type of nozzle. A rectangular apparatus of aspect ratio 3.57 and flexible walls enabled a parametric study of the mean and turbulent properties of the jet plume versus nozzle pressure ratio (from 1.2 to 2.0), exit-to-throat area ratio (from 1.0 to 1.8) and wall divergence angle at the nozzle exit (from 0 to 4 deg.) Time-resolved surveys of total pressure were obtained by means of a dynamic Pitot probe. The growth rate of the jet and the peak rms value of total pressure fluctuation near the nozzle exit increase several fold with area ratio. This trend becomes most pronounced for nozzle pressure ratio around 1.6. At fixed area ratio and nozzle pressure ratio, the wall divergence angle has little effect on the instability. [Preview Abstract] |
Tuesday, November 22, 2005 8:52AM - 9:05AM |
LQ.00005: Crossflow Influence on Transverse Jet Shear Layer Instabilities: Experimental Studies. Sevan Megerian, Juliett Davitian, Ann Karagozian This experimental study examines alterations in the character of the nearfield shear layer instability associated with a gaseous transverse jet, with specific focus on the influence of the crossflow. Two different convergent nozzles of the same shape are utilized: one which is flush-mounted in the injection wall and one which extends from the injection wall, allowing exploration of the effect of the wall boundary layer and associated presence of a horseshoe vortex. A range of jet-to-crossflow velocity ratios ($ 1 \le \rm R < \infty $) and jet Reynolds numbers (1800 $\le \rm Re_j \le $ 3800) is explored. The flush-injected transverse jet undergoes a significant transition in the nature of the shear layer instability as the crossflow magnitude is increased, where very distinct fundamental, harmonic, and (in some cases) subharmonic modes are excited. No significant transition is observed for the jet emanating from the extended nozzle. While these differences in the characteristics may be marginally related to an effective reduction in the crossflow magnitude within the wall boundary layer, there is evidence that it is the presence and behavior of the horseshoe vortex system that significantly influences the flush-injected jet's shear layer instability for $\rm R < 4$. [Preview Abstract] |
Tuesday, November 22, 2005 9:05AM - 9:18AM |
LQ.00006: Crossflow Influence on Transverse Jet Shear Layer Instabilities: Theory and Computations. Leonardo Alves, Robert Kelly, Ann Karagozian Linear stability analysis and 3D transient numerical simulations of the round jet injected normally into a crossflow are utilized to explore the influence of the crossflow on changes in transverse jet nearfield shear layer stability characteristics. The stability analysis explores two alternative continuous base flows at the jet exit, in contrast to earlier discontinuous base flows\footnote{Alves, L., et al., {\it Bull. Amer. Phys. Soc.}, Vol. 49, No. 9, p. 52, 2004.}, and represents the upstream crossflow as spatially uniform, hence representative of a jet nozzle extended from the injection wall. The 3D low Reynolds number numerical simulations, on the other hand, explore the influence of crossflow with an injection wall boundary layer upstream of a flush-mounted transverse jet. Each approach provides support for experimental observations\footnote{Megerian, S., et al., {\it Bull. Amer. Phys. Soc.}, Vol. 5 0, No. 9, 2005.} on differences in the shear layer instability between flush-injected and extended nozzle-injected transverse jets. The results indicate similar trends to those of experiments in the variation of disturbance amplitude and Strouhal number as the jet-to-crossflow velocity ratio is reduced from 10 to 4. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700