OCR Text |
Show Figure 9 shows contours of turbulent kinetic energy and Reynolds stress. The highest values of turbulent kinetic energy and Reynolds stress were seen where the jets entered the duct and near the center of the duct where the opposing jets Interacted. Figure 10 shows a comparison of two test cases where the bulk velocity ratios (VR) were 10 and 20. All test conditions other than the velocity ratio were the same for the two cases. The notable difference between these two flows can easily be seen. For the VR=20 case, the mean flow was symmetrically distributed across the duct, while for the VR=10 case, the mean flow always followed either the top or the bottom wall and hence was significantly asymmetrical. It should be noted that there was no noticeable preference for the flow to follow either the top or bottom wall. A very small perturbation in the jet flow rates would cause the flow to switch from one wall to the other and it occasionally switched spontaneously. The data shows that there was very little decay in the wall-following jet velocity in the downstream direction in the VR=10 case. Furthermore, unlike the VR=20 case, zones of very high turbulence were not found In the flow field for the VR=10 case. Instead, a well-defined shear layer extending far downstream of the jet injection station was indicated. Therefore, it seems safe to conclude that the primary and secondary flows were poorly mixed in the VR=10 case. Another undesirable characteristic of a wall-following flow is the uneven heating load on the combustor walls. For these reasons, a velocity ratio of 20 would be preferable to a velocity ratio of 10 for combustor operation. Figure 11 shows a comparison of two test cases where the jet injection angle was varied from 90° to 120°. All test conditions other than the jet injection angle were the same for the two cases. The behavior of these two flows downstream of the jet injection point was very similar, but the distribution of turbulence around the jets indicated that 120° injection causes higher turbulence near the duct walls while 90° injection causes higher turbulence near the center of the duct. While it would be difficult to conclude that one injection scheme is definitely better than the other, 90° injection would probably be preferred since higher turbulence level regions should correspond to regions of more intense mixing and combustion and it is desirable keep the hottest part of the flow in the center of the duct. 13 |