OCR Text |
Show The initial tests used burners designed for 25 million Btu per hour and an air preheat temperature of 500 deg F. For these tests the first 20 foot of the interior surface of the combustion chamber was lined with refractory fiber to limit heat absorption and provide gas temperatures of 2200 to 2400 deg at the exit of the furnace. Tests were conducted in which the fuel injection parameters (number of ports, location, injection angle, and primary/secondary fuel ratio) and air swirl were varied. N o tlt') o ~ c w ~ () W II: II: o ~ 350- 300- 250- 200 150 ° 4HIgh Spin t:. ., Moderate Spin o 10 Low Spin a: a: ::> 2 cf:---o -d> CD g 100- _---~D---.fo---"tJ 7<.,.-cr J~ ...J W ::> Ii. o W _ ..[J ~ _-0-- -- t- - en g..Q--"""- ________ ~---t:. t:;,-- -- t;. er---- 0 50 0------- 25 M Btu/hr. Ol~-----,~----~----~,----~,~----,-, o 1.0 2.0 3.0 4.0 5.0 OXYGEN IN FLUE GAS, % (DRY BASIS) Figure 4 NO vs. O2: Comparison of Baseline and Sta~ed Fuel Configurations 1400 1 1200- 1 I 1000 I 2 I I Ii. I Ii. I Iff 1 I 4( 1 " 800 I W I .;.:.) 1 I II. I ! 600 \ I 0 I () , I \ \ 400 \ \ ° 2.5% 02 \ \ \ , \ 200 ''2'' , , , '-0-_ 25 M Btu/hr. I I 6112 10'/2 14'/2 18 '/2 22 '12 Stack SAMPLE LOCATION IN FURNACE, FEET FROM BURNER Figure 5 CO vs. Distance from Burner at Moderate Spin for Staged Fuel Burner Configuration 347 Figures 4 and 5 give the results of these tests. Figure 4 shows the NOx emissions of the best staged fuel configuration compared to a conventional burner. Data are given for three levels of swirl. The high swirl adjustment gave an intense, short flame. The NOx emissions were much less than those of the conventional burner. The low swirl adjustment resulted in comparatively longer flames, but the lowest NOx levels. Visible flames generally correspond to the point at which about 99% of the fuel is consumed and at this point carbon monoxide levels are in the range of 1000 to 2000 ppm. Figure 5 shows the results of measurements of CO along the furnace axis at the centerline for the 25 million Btu per hour staged fuel design with moderate swirl. These measurements confirm the visual estimates of flames less than 8 to 10 feet long for the staged fuel design. Based on the success at 25 million Btu per hour, the next step was to design and test a burner for 100 million Btu per hour heat release. The bulk of the tests with this burner were also conducted at air preheat temperatures of about 500 deg F, but additional data were collected at other combustion air temperatures from ambient to 500 deg F. Enough of the ceramic fiber insulation was removed from the furnace cooling surfaces to provide adequate heat absorption and realistic furnace exit gas temperatures in the range of 2200 to 2500 deg F. 350 325 300 N 275 0 C?P?. 250 0 ~ 225 Q w ~ 200 () w ~~ 175 0 () 150 ~ ~ ~ 125 )( 0 z 100 75 50 25 -0-0--0-- High Swirl ,....,.---.::?-Ul'------~ Low and Moderate Swi"1 v -L'lLf L.> _-0 Ful Open (No Swi1) 100 M Btu/hr. 1.0 2.0 3.0 4.0 5.0 OXYGEN % IN FlUE GAS (DRY BASIS) Figure 6 Baseline Configuration: NO vs. Excess Oxygen for Various Spin Vate Positions with 475°F Air Preheat Figures 6 and 7, respectively, show the baseline and staged fuel results at various swirl levels and excess air levels. The baseline data was obtained with a convention'al design 100 million Btu per hour burner and produced NOx emissions in the range of 280 to 305 ppm. This was somewhat higher than that seen with |