OCR Text |
Show FGR AND PREHEATED COMBUSTION AIR FGR is a diluent which contributes substantially more mass per atom of oxygen than does combustion air, and inhibits N O x formation primarily because it lowers the flame temperature. The formation of thermal N O x is inhibited, but not the formation of F B N N O x and prompt N O x which are comparatively low temperature reactions in boiler flames, making them insensitive to the temperature suppression resulting from the addition of any inert diluent. The fact that F G R does not retard the formation of F B N N O x is expected and the data confirms it. The fact that the N O x increases with addition of F G R to high nitrogen fuels is likely not due to the chemical kinetics of nitrogen and oxygen free radicals, but likely due to the increased turbulence leaving the burner register as the register draft loss increases with the increased mass flow through the register at constant heat input and boiler outlet oxygen concentration. When the oxygen concentration leaving the flame is held constant and F G R is added to the air entering the burner, the N O x yield from high nitrogen fuel increases. The greater percentage F G R added to high nitrogen fuel oil combustion air, the greater the N O x increase. The author acknowledges that not all investigators are in agreement that the addition of F GR to high F B N flames increases N O x . The apparent reason that a body of anecdotal evidence is available indicating the reduction in N O x from high bound nitrogen fuel is, that, in practice the simplest test to conduct in an operating plant is to maintain the existing control system characterization while adding FGR. The corrected N Ox is reduced with this method of adding F G R because the oxygen concentration is reduced. Further, reporting N O x as a volumetric concentration with the N Ox measured at a lower oxygen concentration leaving the flame and then adjusting to a regulatory reference oxygen value further reduces the reported NOx. Of course, N O x reported as mass fractions are not subject to volumetric reference corrections. The actual N O x is reduced with this method of adding F G R because the oxygen concentration is reduced as F G R is increased, not because of the flame temperature suppression. Even though N O x is not reduced with the addition of F G R for high nitrogen fuels burned at constant oxygen exiting the flame, it is well documented that F G R does quite substantially reduce N O x from low bound nitrogen fuel flames, as well as from flames devoid of F B N. APPLICABILITY A wide variety of boilers have had flames installed in them which have been designed by the analytical method discussed above. In addition to yielding simultaneously lower C O and N O x with excellent flame fit, other benefits are being realized. Atomizing steam flows have been reduced, recirculated flue gas has been reduced or eliminated, burner diameters have been reduced without increasing burner draft loss for the same gross heat input, and overall steam generation and power generation efficiency has been increased. One of the most successful sites consists of identical, single burner boilers with a total maximum continuous rating of 150,000 pounds per hour of superheated steam each. The flame shaping technology applied at this site solved several preexisting problems unresolved by the original manufacturer. In addition, it considerably improved plant operability and efficiency. T w o boilers were retrofitted with asymmetric gas flames and symmetrical oil flames. The burner flames were designed by the analytical methods which solved all technical problems related to combustion and flames. In addition, the burners with flames designed by analysis simplified the plant and improved its overall efficiency. The benefits demonstrated at this site are; precise flame shape and fit, correct superheater temperature, low C O , low opacity, reduced atomizing steam consumption, low N O x which is compliant without any of the 20 percent recirculated flue gas originally included in the plant design, elimination of the recirculated flue gas fan, 30 percent reduction in forced draft fan power consumption and complete flow system dynamic stability. The plant has no operational limitations related to combustion and flames. It is now capable of operation at maximum continuous rating with a motor power savings of approximately 100 horsepower per boiler and only one-eighth of the typical atomizing steam flow. Total plant steam generation efficiency improvement directly attributable to burner and flame shape optimization was 1.2 percent. CONCLUSION The analytical method of designing flame shapes has been validated in the full scale Advanced Combustion Research and Development Centre, Hamworthy Combustion Engineering, Ltd., Poole, Dorset, England. The, more than one-hundred hours of, validation tests were conducted exclusively using hardware designed by applying the analytical methods desired to be evaluated. All of the designs were completed before testing began, and all test hardware was manufactured from those design analyses in advance of testing. Measurement instrumentation was calibrated both before and after testing using calibration gases with specie concentrations which were nearest the actual flue |