| OCR Text |
Show 4 The heat balance used for approximating the flame temperature, T~, ignores radiation from the flame itself and assumes that the burner surfaces have an effective emissivity equal to 1. The total radiant heat flux is taken as the radiant heat transfer of the total surface area of the burner, that is, of both the burning and the regenerating surfaces added together. This was justified on the basis that when the chamber wall temperature above the burner segments and the load temperature were stabilized the surface temperature of the burner segments did not visibly change very much as the RRB was cycled. It was observed that once the system stabilized at normal, high operating temperature there was no visible flame, and it was very difficult to tell by eye which of the two segments was burning as they were cycled. The operation ·in that sense was likened by one observer to a reactor rather than a burner. Referring to Table I the values at time 11:42 were typical of at least 1~ hours of operation. The other data points were taken at variations made in flows to explore the limits of operating in terms of NOx, CO and unburned hydrocarbons, O2 level, diluent flow, and fuel flow. The 2:50 test was atypical in that the fuel flow was much higher than for the rest of the points. It showed some of the combustion was probably outside of the burner (evidenced in part by the relatively high CO emissions). The 3:00 point was for very high excess air. Although the NOx level at the operating point (air flow at 5~ times the stoichiometric air) was reasonable, 10 ppm, the equivalent value referred to 3% oxygen is 48 ppm. Despite this, the test showed that air could be accepted as a reasonably effective diluent for NOx control providing NOx emissions limits were expressed as mass of NOx per unit heat generation. Figure 3 shows log NOx vs liT, where T is the calculated flame temperature in degrees Rankine. The correlation is quite wor}cable except for the very high excess air point. The data at various operating conditions is not VOluminous enough to correlate more exactly with oxygen level and flame temperature. Implications The MKE RRB concept - the Pure Hea~M burner - has shown promise of being a new class of high efficiency., very low NOx burner suitable for a wide variety of industrial uses. It should be found useful in the evermore competitive industrial world to come. |
