OCR Text |
Show Gas comp?sition and temperature f!1easur~ments were also made in-flame using intrusive extractive gas samphng probes. Advanced diagnostics were used to characterize the flame structure and to help defi~e mixing behavior and reacti<?n zO,nes. These included use of laser doppler veloci~etry to c.haractenze the ~ow field, ~CH che~luml~escence to identify regions of fuel consumption and hkely NO formation, and Mle scattenng to Illustrate the gross features of the flow field and fluid mixing behavior. The diagnostic plan used is outlined in Table 2. Table 2. Diagnostic plan for testing the Selas K988 burner at the SERL. Diagnostic Rame No. 1 2 3 4 5 A laser velocimetry V' V' V' V' V' CH chemiluminescence V' v' V' V' in-flame extractive gas sampling V' V' V' V' instantaneous planar mie V' V' V' scattering near-field temperature V' V' V' V' measurements RESULTS The control and repeatability of the testing was highly precise. Repeated baseline tests achieved precision at: 0.27 percent in fuel flows, 0.25 percent in furnace gas temperatures, 6.2 percent in radiant heat flux, 1.1 percent in water heat extraction and 2.7 percent in stack NOx emissions. The BERL provides means for precise metering of process streams and for the diversion of air and fuel to various lines. The facility is equipped to supply preheated combustion air with temperatures up to 2000°F and to recirculate high temperature (I 600°F) and low temperature (300°F) flue gas to the burner. The Selas burner study required fine process control of fuel and air, and nitrogen diluent to various burner registers. . Burner characterizations were conducted by varying overall furnace stoichiometry, gaseous fuel formulations, secondary air to fuel volume ratio, secondary jet velocity and fuel staging ratio. Overall furnace stoichiometry, effected by the input of tertiary air, caused a familiar impact on flue gas temperature and NOx and CO emissions. Figure 3 illustrates the impact of overall stoichiometry. NOx emissions decrease with decreasing stoichiometry, however, as the burner operation approaches stoichiometric conditions, CO emission rise dramatically. The furnace gas temperature was measured two feet axially downstream of the burner. As expected, highest temperatures were achieved at stoichiometric operation. Investigation of Fuel Composition Impacts: Natural gas was utilized as the baseline fuel for this study. In order to assess the impact of fuel composition on NOx emissions, independent of other experimental variables, mixtures of hydrogen and propane were used to formulate alternate fuels. The use of alternate fuels simulated the firing of refinery fuel gas. All fuels had consistent heating values and the burner heat release rate and furnace gas temperatures were maintained. In effect, 6 |