| OCR Text |
Show the natural gas/air cases. In general, the CO emissions for the opposed-frring cases were less than 35 ppmd. Upon comparing as a group, the cases in which the fuel and oxidant were delivered into the furnace from the same face (see Fig. 8) to those cases in which the reactants were injected opposed to each other (see Fig. 9), the NOx emissions were lowest for the former group (co-frring). In addition, all of the natural gas/air cases exhibited very low NOx emissions. The data in Figs. 8 and 9 are plotted on the same scale, and the co-frring NOx levels. are often a factor of two lower than the opposed-frring cases. Low emission levels for the arr cases are due to the fact that the oxygen is already significantly diluted with nitrogen prior to entering the furnace in contrast to the case when pure oxygen is directly introduced into the furnace. In both co-frring and opposed-frring cases, some oscillations in the furnace pressure were observed, especially at low furnace temperatures. It is believed that the cause of these oscillations are related to the high flame lift -off distance under high fuel injection velocity conditions, and to the mixing conditions of the fuel and oxidant within the given furnace geometry. No such oscillations were observed with multiple fuel and oxygen lances installed in commercial glass furnaces. Flame stabilization techniques such as an annulus oxygen flow around the fuel jet may be used to anchor the flame to increase flame stability [1], if necessary. 10-2 _ .... - _ ... "0 E Cl. Cl. Furnace Temperature/Nitrogen Content NOx formation is strongly dependent on temperature and nitrogen availability [26]. A series of tests were undertaken to establish the NO x emission characteristics of the DOC system 10' ~ as a function of furnace nitrogen content and ...., "0 ~ "- en 10-3 0' furnace temperature (see Fig. 10). The burner 0 z a:: 1 00 ~ arrangement selected for this test series was a : co-frring arrangement. The frring rate was 10-4 o 1. 3 ~ H2 6 6 . 4 ~ Hz o 13~ H2 G) ·0 10- ' '-". o z o 4 2~ H2 "7 S 4 ~ H2 • 77'7. H2 1 0 -~ L--...~...I--~-'-----~-J.-~.........L.....~...-..J 1100 1200 1300 1400 1500 1600 TwO K Fig. 10. The power emissions index, PINox, as a function of furnace nitrogen content and furnace wall temperature, T w. A co-frring burner arrangement was employed and the nominal frring rate was 185 kW. 185 kW, the furnace wall temperature was varied from 1100 to 1550 K, the furnace nitrogen content was varied from 1 to 77% (vol. wet), and the oxygen concentration in the flue was held to a nominal value of 2.5 % (vol. wet). As shown in Fig. 10, NOx increases with increasing nitrogen content and furnace wall tem~;rature, as expected . . NOx emissions under 5·10 g/MJ (-10 ppmd arr equiv. @ 3% 0 dry) were acquired throughout the entire tes~ temperature range for furnace nitrogen levels under 40 %. For several of the low nitrogen levels (1 .3 and 6.4%), the NOx apparentl decreased with an increase in furnace wait 13 |
