OCR Text |
Show 8. As a result of the above accepted conditions for the calculations, 6 % EFGR provides a sufficient N O x reduction in the C F G flame (Fig.7) and a limited one in the injector flame (Fig.8). The total N O x reduction (from - 38 to - 32 ppm) is shown in Fig.9. It means that 1 % EFGR reduces N O x of - 2.5%. If we suggest that in the EFGR range from 0 to 1 2 % we have a linear dependence of N O x reduction on EFGR flow, increasing the IFGR flow from 6 to 8 and 12 % will bring N O x reduction to -29 and - 26 ppm, respectively. 9. Unlike NO, it is very hard to reduce NO2 significantly by combustion method implementation. As per the accepted test data, there is known 10 to 15% NO2 reduction only. It means that if w e suppress N O in the C F G flame from 97 to 25 ppm (75% reduction) and in the injector flame from - 20 to 16 ppm (20% reduction), at the limiting conditions, we will get 16.7 p p m uncontrolled N O only. At the same time we will obtain unsuppressed NO2-- 14.1 ppm. So, taking the considered data as actual one, we cannot get uncontrolled N O x concentration less than - 30.8 ppm (it is uncorrected concentration). Respectively, this data correction @ 3 % O2 lets us to get numbers <30 ppm. We understand that the above accepted conditions could decrease errors of the calculations but could not bring them to zero, that is why, unlike of the test data, calculation results have allowed us to analyze a conditional process and to get some information that is inaccessible during the test. By the way, the calculated N O x and C O concentrations are close to the test data obtained at the boiler facilities. Based on the performed calculations and analysis described above, it is suggested that the Variflame-2 burner will be modified with changing design of the above mentioned subassemblies. The above modification will provide both (1) better performance data and (2) manufacturing cost reduction. |