OCR Text |
Show 6. The combustion chamber pressure, maintained very slightly positive (10 to 20 Pa) with respect to the ambient air, in order to prevent any infiltration. The principal observations were of: 1. Exhaust-gas composition in respect to O2, CO2, C O , CH4, N O and NO2. Other work in the C A G C T laboratories (Turnbull 1995) has shown that the N O 2 / N O ratio is significantly affected by probe variables and, because of this, it is difficult, perhaps impossible, to reliably resolve these two species. Their sum, expressed as "NOx", is, however, remarkably consistent. W e therefore report N O x but do not attempt to distinguish between N O and N O 2. 2. Refractory-surface temperature distribution 3. Sink heat flux distribution. 4. Combustion-chamber static pressure fluctuation 5. Flame and refractory appearance 6. Manifestations of combustion stability/instability (magnitude of pressure fluctuations, flame visibility and stability, need for continuous pilot flame, etc.) 7. In the fourth series of trials, the gas composition field in the combustion zone of burner Bl. The work was begun with a definite plan for a test matrix. However, growing experience with the burner, tempered by accidents and discoveries along the way and by the considerable costs of the trials, had a determining effect on the pattern that evolved. A s already noted, the conclusion of the work, fortunately only after substantial results had been obtained, was forced by the corrosion of the mild-steel burner bodies. A summary of the configurations of the furnace chamber, the heat sink and the burners in the trials that were conducted is given in Table 4. It may be noted that only one fuel-port diameter, D\ = 6.35 m m , was employed. This is not a serious deficiency in the experimental design, since the most critical operating parameter associated with the port sizes is actually the fuel/air momentum flux ratio, y/u, which was significantly varied. The tunnel configuration for the Series 1-XBM2-T trials is illustrated in Fig. 3. The tunnels used in the Series 1-XBM1-T/F trials, described in detail by Becker and Sobiesiak (1996), were fairly similar to this. Figure 3, incidentally, also shows the layout of the water-cooled floor panels constituting the heat sink or "furnace load". The sink configurations used to vary the furnace load in the Series 3-XBM2-F trials, with various coverings of the floor panels with blanket insulation, are shown in Fig. 4. 7. Results 7.1. Introduction The concern in the research was first with NOx emission, second with combustion stability, and third with flame size and combustion intensity. The results are accordingly presented under headings.. The contract report on the research (Becker & Sobiesiak 1996) includes a full consideration of energy balances, with energy budgets and calculations of furnace efficiency. The productive part of a trial began as steady state was approached. The effectively full attainment of 7 |