| OCR Text |
Show Measurement of each velocity component was conducted separately. Optical system reconfiguration was required for moving between axial and radiaVtangential component measurements. Both radial and tangential components used the same optical setup. A change from measurement of one component to the other was accomplished by rotating the furnace through 90· and scanning the flow orthogonally to the previous scanning direction. In this way all three components were measured across the same radius of the furnace. It should be noted in Table II that the probe volume used in this study was measured to have a diameter of 130 Jlm (1/el) and a calculated length (as imaged) approaching 2 mm. This probe volume is somewhat longer than optimum, but is the minimum that could be obtained given c~nstraints of collection optics, particle scattering cross-section, laser power, and required Stgnal-to-noise ratio. In order to minimize possible gradient smearing effects, the probe volume was oriented with its long dimension along the tangential coordinate in the furnace (no mean gradients) for axial and radial component measurements. This could not be done for tangential component measurement, and this effect must be kept in mind when viewing the data. RESULTS Figure 3 shows a combined schlieren and luminosity photograph of the region of the flame under study. This photograph was obtained using a conventional laser schlieren optical arrangement (see, for example, Keller and Saito, 1987) but collecting flame luminosity around the outer edge of the schlieren aperture. The dense spray region is apparent as a dark "V" shape at the . centerline of the nozzle. The atomizing air also enters along this path. The main combustion au supply enters through the furnace throat immediately adjacent to (concentric with) the nozzle. The dark region originating at the outer edge of the furnace throat corresponds to a sharp density gradient between the cool incoming air and the hot burned gases of the furnace. This boundary curves outwards as the main air stream spreads. The bright upper region of the photograph corresponds to strong flame luminosity. This marks the beginning of a zone of high combustion intensity (starting at about z = 100 nun) which persists for about one-third of the furnace height. A small tongue of flame is visible within the spray cone. This internal flame indicates the presence of a strong central recirculation zone-observed previously for swirl numbers greater than 0.6 (Beer and Chigier, 1972, Gupta et al. 1984). The dark regions near r = 100 mm, z = 100 nun, are from density gradients along the innennost edge of a torroidal vonex positioned adjacent to the furnace windows and floor. Not evident in the photograph are small flamelets (-5 nun in width) which appear intermittently on the outer edges of the spray sheath. These flamelets originate in the luminous flame region and are convected downwards (toward the nozzle) along the outer edge of the spray while simultaneously circling the periphery of the spray sheath. After spiraling down to as far as z = 25 mm these flamelets are convected downstream and disappear. Figure 3 shows the region of the flow in which the velocity data were collected. For presentation, the data will be split into two parts. In the first part, data is presented throughout the flowfield in a manner which facilitates comparison with Figure 3, and in doing so, illustrates the structure of the flowfield. The second gives details of the inlet air flow (z = 0 mm plane) and is intended to assist those interested in modeling this type of flow by providing well defined inlet conditions. -4- |
