OCR Text |
Show interaction between the throat velocity and premixing distance. The only factor which does not have an effect on N O x emissions is the jet velocity. The reason for this m a y be that the jet velocities investigated all impinge on the annulus wall, thereby providing insignificant differences between cases. The C O emissions are only affected by the throat velocity and an interaction between the throat velocity and swirl input. The fact that so many parameters have an effect on the N O x production is a testament to the complexity of the problem. Based on these results, a model was developed in DesignExpert to extrapolate over the space encompassed by the factors. This model was used to plot the N O x and C O emissions contours in terms of any of the given factors in order to determine trends. These plots are provided in Figures 5 - 7. In Figure 5, N O x and C O are plotted in terms of the throat velocity factor and swirl input. For low N O x , higher throat velocities and higher swirl values are recommended. The higher throat velocities provide better cross-flow mixing and a more diffuse jet wake prior to ignition. The higher velocities reduce the residence time in the throat and increase the shear on the ignition and reaction zones. Increasing swirl levels produces higher levels of internal FGR. These effects all collaborate to reduce N O x emissions, but can also increase C O if quenching occurs rather than cooling. The C O plot illustrates that swirl is a more dominant factor than throat velocity, as indicated by the nearly vertical contours for S' < 0.58. The N O x and C O response contours as functions of the number of jets and throat velocity are shown in Figure 6. As established in the previous figure, low N O x favors higher throat velocities; this figure shows that, in addition, a lower number of fuel jets leads to lower N O x. The use of fewer fuel jets would tend to increase the spacing between the jets such that adjacent jet interaction and the number of ignition sites are reduced, thereby reducing the number of low velocity, high temperature, rich regions. The larger distance between jets would also allow increased cross-flow mixing and air interaction, creating more diffuse ignition sites between jets. The C O emissions are not affected by the number of jets. The premixing distance effect on N O x and C O is shown in Figure 7. The C O emissions do not change while the N O x emissions increase with increased premixing distance, and the effect is more pronounced with lower velocity factors. This is counter to the conventional hypothesis that N O x should decrease with higher levels of premixing. These results indicate that the degree of premixing is not the only mechanism being addressed by this factor. |