Meeting SCAQMD Rules 1109 and 1146 with Low NOx Burners

Page 5

AFRe Fall 1990 water circulates through the convection section coils of the heater. The heater radiant section measures 12'-8" in diameter and is 33'-9" high. The nominal bridgewall temperature is about 12000p with an available draft at the floor of 0.4 inches w.c. The heater was fired on refinery fuel gas with a heat content of about 1350 B tul scf for all reported data. The NOx control steam injection supply was regulated down from a 150 psig header to 80 psig for the high steam flow case and 40 psig for the low flow case. This arrangement provided for a dry steam supply to the burners. Condensate could collect in the piping during periods when the steam flow was not in use and caution was used in initiating steam flow to the burners to minjmize liquid water injection into the combustion zone. After starting up the heater with the new NDR burners, data were collected operating with and without steam injection. The initial data were collected with a portable emission analyzer (ENERAC 2000) and later with a chemiluminescent analyzer. The flue gas sample was drawn from the top of the firebox at the entry to the convection section. After analyzing the test data it was concluded that air leakage into the radiant section varied with the amount of draft at the time of the test. In order to put all of the test data on the same basis the actual flue gas measurements are shown as well as adjusted curves which take into account the estimated leakage. This adjustment was determined by the excess oxygen ~ead.ing at which the heater showeclsigns of running out of air or smoking. Figure 5 sho"WS NOx data collected with the portable NO meter for the unit operating at I the normal input of about 28 :MM:Btu/hr wi~ a box temperature of about 1200Op. Figure Sa shows the same curves corrected for the estimated leakage. The upper curve gives NOx versus excess oxygen without steam injection and the lower curve gives the results when approximately 0.15 lb of steam per lb of fuel were added. Both sets of data show a relatively linear relationship between NOx and excess oxygen. In addition the slopes are nearly identical. Although this furnace is subject to the NOx limit prescribed by Rule 1146, the data show that the burners would meet the Rule 1109 limit (approx. 26 PPM) with steam injection if the excess oxygen at the burners were controlled to under 4%. Figures 6 and 6a show NOx data collected with the unit operating at maximum fIring rate. In this case the furnace temperature has increased and the NOx levels have increased as well. For this condition it would be necessary to operate the furnace with steam injection and less than 3.5% excess oxygen at the btp"TIer in order to meet Rule 1109. Figures 7 and 7 a ~how data collected on this furnace with a ThermoElectron Chemiluminescent NOx analyzer. ~pata are given for 3 steam injection rates which range from about 0.06 to about 0.12 lb of steam per lb of fuel. Interestingly, the data in this test showed almost no relationship between NOx and steam flow over this range. The data, however, does confmn the ability of the burners to meet the Rule 1109 limit at excess oxygen levels of about 2.5% at the burners corresponding to about 4% 02 at the test point at the top of the radiant section. 4