| OCR Text |
Show is near constant averaging about 14 ppm. This is almost identical to the results shown in Figure 4 for a 15 k W system despite a forty times difference in input power. The highest N O x level is found for the 10 c m L S B at 585 k W (21 ppm). This is the condition for which the swirl air supply was limited and it was not possible to operate at a more stable (i.e. slightly higher) swirl number. Figure 6 shows that it is very close to the blow-off limit. Therefore, the slight increase in N O x may be a direct effect of this near limit operating condition. The corresponding CO emissions shown in Figure 8 demonstrate a very strong dependence on the input power as well as on the chamber size. Differences in C O emissions measured in two configurations with an almost identical 5 c m L S B clearly illustrates the significance of the burner-chamber coupling effects. At 18 k W the L B NL water heater simulator (Ac / A b = 15) achieved C O = 50 ppm. But the same burner operating at the same input power in the burner evaluation facility (Ac / Ab = 142) produced C O emissions approaching 2500 ppm. At higher input powers, C O emissions decrease drastically and level off to about 25 ppm. Note that the straight lines through the data serve to guide the eye, but they are not mathematical fits. The turning points of each set of data provide convenient markers for the minimum input powers to ensure low C O operation. For the 5 c m burner installed in the burner evaluation facility (Ac / Ab = 142), it needs to be fired higher than 65 k W . In the furnace simulator, the minimum input power would be 400 k W . By extrapolating the data for the L B N L water heater simulator, it seems that a minimum input power of 25 k W would be sufficient to burn out the CO. In Figure 9, the data obtained for unburned hydrocarbons also show strong burner-chamber coupling effects. The exceedingly high U H C emission (2800 p m ) at the lowest input power of the small L S B (17.5 k W ) indicates very poor combustion efficiency. This is consistent with the corresponding high C O emissions shown in Figure 8. Such poor performance is perhaps a consequence of a small flame burning within a large combustion chamber. Interaction with the cooler air that surrounds the flame can quench the combustion reactions (hence high U H C ) and cool down the products to prevent C O burn-out (hence high C O ) . The rapid decreases in both U H C and C O emissions with increasing input power seem to support this conjecture. Doubling the input power of the 5cm L S B from 18 k W to 36 k W , U H C production is reduced from 2800 p p m to only 60 ppm. A similar trend is also shown by the set of data obtained for the 10 c m L S B where U H C emissions drop to undetectable level at about 400 k W . These test results clearly illustrate that combustion efficiency and C O emission are system optimization issues because they are associated with burner-chamber coupling. However, they also show that in a L S B system N O x emission is not an issue because it is independent of this coupling. In California, development of stationary heating and power generation equipment has been driven primarily by the increasingly strict rules adopted by the Air Quality Management Districts ( A Q M D ) . With over half of California's population, the South Coast A Q M D and the Bay Area A Q M D have been developing new regulations to reduce pollution within the district boundaries. Emission controls imposed in these two regions often become benchmark standards for the nation. N O x is among the most regulated 8 |
