OCR Text |
Show 8.4.5. Effects of pilot burners W e note again that having the pilot burners on or off had no discernible effect on the N O x emissions, even though combustion stability was sometimes strongly affected. 8.5. Combustion stability While trials that produced useful data were achieved with steady-state exhaust-gas temperatures as low as 900 °C and average refractory temperatures as low as 830 °C, operation at such low temperature levels is at the brink of general instability and cannot be recommended. Based on the present evidence, w e advise against operation with Te < 1000 °C or Tr < 900 °C. For best performance, it seems advisable to aim for Te > 1050 °C and Tr > 930 °C. The level of excess air required to maintain stability, as illustrated by Table 7, varies with the exhaust gas temperature and with the firing rate. At very high gas temperatures, say Te > 1700 °C, combustion at high fire is very stable even under markedly fuel-rich conditions. At the very low temperature Te = 900 °C, at least 15 % excess air is required (XQ = 3 % ) . Except at very high temperatures, but with Te > 1050 °C, it seems advisable to provide at least 10 % excess air (ca. X0 = 2 % ) . This should also ensure stability at turndowns of at least 3:1 (reduction of firing rate to 33 % of full fire), if not more. W e note again the absence, within the practically interesting operating bounds of combustion air pressure, of an upper stability limit on the firing rate. The present work was done with a system that limited the pressure drop across the burners to about 5000 Pa, and this determined the high-fire limit of operation. Use of an air supply providing higher pressures would increase the available m a x i m u m firing rate and correspondingly increase the available turndown ratio. The fuel jet angle strongly affects stability if too large. Generally speaking, the smaller J3\2 is, the sooner will fuel and air meet, and the shorter will be the flame, all enhancing stability. Of course, if fin is decreased too far, the fuel and air jets will entrain too little of the recirculating product gases before meeting, and the N O x will begin to rise. There is some indication in our results of this happening in going from 0X = 35° (J3n = 26.3°) to 30° (21.4°). At the clearly excessive fuel jet angle 0\ = 65° (fi\2 = 56.1°), at which impingement of the fuel jets and of the combustion zone or "flame" upon furnace boundaries was very strong, combustion stability was somewhat shaky at all temperature levels. At high exhaust temperatures Te, behaviour was seen in the trials with tunnel combustion chambers that did not occur at smaller 0\. The gases in the combustion zone, which almost filled the combustion chamber, were sometimes laden with particles, evidently soot, giving the appearance of a blizzard in lamplight. Heavy pressure oscillations usually occurred at the same time. Based on experience to date, w e recommend that burners be operated with a continuous pilot flame. Since a pilot is needed anyway, for startup, there is no significant cost for this. The N O x emissions are not perceptibly affected. Clearly, the design approach taken in the C G R I burner which produces such very low N O x emissions inherantly carries a penalty on stability, and the continuous pilot is perhaps the price that as to be paid. Perhaps a pilot used in this manner should not be called a pilot but something else, say an auxilliary burner. 23 |