OCR Text |
Show enter the combustor, such that velocities and turbulence intensities were quite similar across the bulk of the flow. For this reason succeeding plots show data for only the plane bisecting a jet. The flow was found to have a periodic wake regime in all cases studied. Figure 7 shows a vertical velocity-time history at a point on the centerline of the duct for the case plotted in Figure 5 as well as a frequency spectrum plot which defines the dominant frequency of the flow oscillations. Observation of the flow indicated that the periodic wake regime resulted from the alternate growth and shedding of vortices produced by interaction of the jets and the bulk flow. Primary frequencies ranging from 0.7 to 1.6 Hz were observed and a dependence on the duct area reduction was indicated, but no trend could be established when comparing cases where the area reduction was varied. In the wake regime the time averaged data encompasses this periodic vortex shedding and therefore is not representative of a typical flow pattern. The remainder of the field is a stationary, non-periodic turbulent flow, so there the time averaged data is representative of the typical flow pattern. 0 00 2 00 * 00 e 00 BOO 10.00 0.00 8.00 18 00 ACQUISITION TIME (8EC) FREQUENCY (Hz) Figure 7. Vertical Velocity Component Variations in Time at X=l, Y=0, Z=0. Figure 8 shows profiles of mean velocity and turbulence intensity. The bulk velocity of the incoming primary flow (0.092 m/s) was used to normalize the data. The highest turbulence level was found near the jets, and the turbulence intensity profiles became uniform as the flow moved downstream. Mean velocity measurements obtained above the duct centerline are plotted to illustrate the flow symmetry for this case. 11 |