OCR Text |
Show RESULTS 50,000 Btu/hr Burner The objective of the initial exploratory tests perfonned with the sub-scale burner experiment was to explore the hypothesis that improvements in fuel and air contacting and mixing can reduce NOx emissions. The tests were performed by systematically varying burner operating parameters (e.g., fuel injection angle, axial air velocity, and swirl), and determining the effects of these variations on global burner petformance. The baseline operating conditions for the sub-scale burner are axial injection of the fuel at a velocity of 52 ft/s and injection of two equal non-swirling air flows coaxially around the fuel injector at a primary air velocity (VI) and secondary air velocity (V2) of approximately 14 ft/s. At these conditions, the burner produces a classical axial turbulent diffusion flame with a relatively long length. Although flames of this type are used in some industrial processes, this flame does not represent a flame typically used in small industrial boilers. In the present case, this condition simply served as a convenient starting point for the screening since the structure and characteristics of this flame have been studied in detail. At the baseline conditions, the flame length was approximately 29 to 34 inches long and the flame NOx, CO, and hydrocarbon emissions were, respectively, 129 ppm (corrected to 3% 02), 2,229 ppm (corrected to 3% 02) and 111 ppm (corrected to 3% 02). The combustion efficiency of the baseline flame based upon the CO and unburned hydrocarbons emissions was 99.27 percent. (The NOx emissions levels are lower and the CO and hydrocarbon emissions are higher than might be expected in comparison to industrial combustion systems due to the relatively cool envirorunent surrounding the flame.) Selected results of the screening studies are shown in Figure 4. In this graph, the NO emissions are reported as ppm, corrected to 3% 02. The combustion efficiency for each flame is also shown. Figure 4 shows the effect of varying the injection angle of the fuel (00 = axial injection) on NO emissions for two primary ldilution air velocity ratios (V I = 14 ft/s, V 2 = 14 ft/s; VI = 26 ft/s, V 2 = o ft/s). For both of these velocity ratios, swirl was added to the primary air stream. In comparison to the baseline case, the results shown in Figure 4 indicate that adding swirl and increasing the momentum of the primary air stream reduces NO emissions. Visually, these changes also result in producing higher intensity, shorter flames in comparison to the baseline flame suggesting that the fuel and air mixing were improved as well. Figure 4 also shows that substantial reductions in NO emission can also be achieved by varying the angle of injection of the fuel. Visual observations indicated that injecting the fuel with a radial component increases the rate at which fuel and air are brought into contact in the flame. The combustion efficiencies shown in Figure 4 reveal that, under certain conditions, NO reductions can be accompanied by a decrease in the combustion efficiency. Although no attempt was made in these studies to optimize the mixing patterns at these conditions, the results suggest that the combustion efficiency could be increased by tuning of the fuel and air mixing patterns. These results indicate that fuel and air contacting and mixing can significantly affect the emission 6 |