OCR Text |
Show Caution is advised when extrapolating these results to more than the MIT Combustion Research Facility test furnace. There were many concerns with the RSFC burner. One concern was that the 10w-NOx configuration of the RSFC burner (RSFC-A) included the need to retract the coal nozzle about six inches into the primary air zone. However, this caused a glowing red exterior. Further testing with thermocouple readings was determined to be required. Secondly, the NOx emission reductions were dependent on the overfrre air port location in the CRF for all of the burner configurations tested. The very low emission results required long residence times from the burner to the overfrre air port. These residence times would not be feasible in a commercial unit. The MIT furnace was a long, narrow tunnel, and included the use of a single burner. Further testing was determined to be needed to understand the performance effects of frring the RSFC in a vertical furnace shaft, with multiple opposed frred burners. Noticeable changes were found in comparing the operation of the DRB-XCL ® in the CRF and SBS facilities. Finally, the RSFC burner is not a scaled "down version of a commercial product as the DRB-XCL ® burners that were tested. Therefore, many concerns arose in thinking about translating the concept into a feasible commercial offering. These concerns included achieving the high swirl values with reasonable burner pressure drops in an acceptable burner size; resolving the problems apparent in the retracted coal nozzle requirement; and simplifying the air delivery system. Numerical modeling also played a key role in the investigation of an advanced 10w-NOx burner design. Performance evaluations based on flow and combustion modeling of the MIT-RSFC burner, in both staged and unstaged operation, as well as the DRB-XCL ® burner in un staged operation were completed. Predictions from the DRB-XCL ® burner were used as a baseline for comparison with the RSFC. These results were useful in providing additional understanding of the RSFC burner operation and insight into burner development and scale-up efforts in Phase II. In addition, predictions were compared with collected in-flame and exit data for both burners to provide additional model validation before proceeding with Phase II modeling efforts. The numerical predictions compared qualitatively with the CRF measurements for both burners. The modeling was able to predict major trends for temperature and species variations both at the exit and within the furnace. A comparison summary of measured and predicted exit conditions for both burners is given in Table 2. This table shows that the model was able to provide reasonable agreement with measured data, and in particular was able to capture the variation in NOx levels measured between the burners. A comparison of axial temperature and major species distributions along the burner centerline 8 |