OCR Text |
Show 5 A parametric study was carried out using the flame described above as a base case. The effect of transport air to coal mass ratio (f/C) was determined by increasing the fraction of transport air. The primary or tertiary air flows were adjusted accordingly to keep a constant burner stoichiometric ratio. The exit NOx concentrations were measured after reaching steady state conditions in each case. With no compensation on the burner air, the exit NOx concentration increased from about 105 to 150 ppm, when the TIC was increased from 1.4 to 1.8. The higher NOx concentrations are due to the higher burner stoichiometric ratio. This considerable increase of NOx concentration (50%) indicates that the burner stoichiometric ratio has a very strong impact on NOx formation. When the tertiary air flow was adjusted (reduced) to compensate for the additional transport air, the exit NOx concentration decreased from about 105 ppm to 80 - 85 ppm. These results indicate that NOx exit concen~tions are very sensitive to the mass flow rate of tertiary air. On the other hand, when the compensation was made on the primary air flow, a negligible difference was measured in the exit NOx concentration. There was little change in the flame shape during these experiments, even when the primary to transport air mass ratio was changed significantly, indicating stable flame characteristics. The available residence time for the fuel nitrogen conversion in the fuel rich zone is an important design parameter for low NOx boilers. The effect of the first stage residence time was investigated by injection of air at two separate axial positions -10.5 and 13.5 burner diameters downstream of the flame. By changing the distribution of air between the injection ports, the mean gas residence time in the first stage of the combustion chamber is easily varied. The maximum variation of residence time achievable with this configuration is estimated at about 600 msec. During these experiments, the over fire air split between the two injection stations was varied for 5 different cases and the exit gas composition was determined. Introducing all the over fire air at the upstream axial location caused the exit NOx concentration to increase from 110 ppm to 200 ppm. The correlation between the two variables is linear. This experiment underscores the importance of the mean gas residence time in the fuel rich zone in obtaining low NOx emission. The effect of the thermal input was investigated by increasing the coal flow rates while maintaining a constant (transport + primary air)/coal mass ratio. The fraction of tertiary air was allowed to change. The burner SR decreased from 0.92 to 0.82 when the thermal input was increased from 1 MW to 1.8 MW. The variation of exit NOx as a function of thermal load and burner stoichiometry is shown in Figure 3. At lower load (l MW) the burner stoichiometric ratio was 0.92 and at this condition, the exit NOx concentration,was 94 ppm corrected to 3% ~. At maximum load (l.8 MW), the NOx concentration decreased to about 70 ppm, but the stoichiometric ratio was also lower (0.82). Thus, the exit NOx concentration is relatively insensitive to thermal load within the range of investigation. The exit ~ concentration was also varied between 2.0 and 4.5%, both at the staged (SR=0.92) and DOn-staged conditions. In the non-staged flame, the exit O2 concentration was varied by increasing the fraction |