Particle image velocimetry and OH planar laser induced fluorescence are used to examine the flow and flame structure resulting from two adjacent fuel/air nozzles. The distance between nozzles is varied from 1.1 to 2.72 nozzle diameters to change the degree of interaction between the nozzles. Non-reacting PIV shows a flowfield which is nearly symmetric between nozzles for all four spacings. For all but the widest spacing, there exist differences in the flow structures between the inner and outer sides of the nozzles. Less distance between the nozzles results in more rapid merging of the shear layers and higher axial velocities between the nozzles. At a spacing of 1.89 nozzle diameters, the shear layers toward the adjacent nozzle are intermittently pulled into the center of the combustor, resulting in a wider and lower velocity average flow between nozzles compared to the other cases. When a flame is added, the flowfields become much more asymmetric, both between nozzles and between the shear layers toward and away from the interacting nozzle. The outer shear layers, away from the other nozzle, are pushed to the domeplate by the expanding recirculation region. With a higher airflow, this behavior is negated. At the two further spacings, the shear layers toward the adjacent nozzle also become different between swirlers. The flow from the nozzle located on the right stays near the domeplate and joins the shear layer from the left nozzle near the nozzle lip and becomes a strong and penetrating jet at an angle to the axial coordinate. OH PLIF imaging shows that the flame fronts are generally located in the shear layer between the incoming reactants and the recirculating combustion products. This includes cases where the flow is highly asymmetric and stays close to the domeplate or penetrates deep into the combustor. So there is no evidence that the addition of an adjacent nozzle has an effect on the local flow/flame interaction. However, it is also clear that the presence of interacting swirling flows with combustion can lead to very dramatic changes to the global flow behavior relative to a single nozzle experiment.

This content is only available via PDF.
You do not currently have access to this content.