OCR Text |
Show impingement of jet AS on the glass surface causes an increase of maximum breast wall temperatures by 40 R (22 K) compared to the baseline 02 case. Effect of Variable Glass Surface Temperatures The preceding results, which were obtained using a uniform effective surface temperature of the glass, showed relatively large inhomogeneities in the distribution of net heat fluxes at the glass surface. The largest and smallest net heat fluxes predicted for air combustion (Case 1) were for instance 95 kw/m2 and 35 kW/m2 , respectively and were 110 kW/m2 and 40 kw/m2 for the baseline 02 case (Case 3). For the 02-cases in particular, the local heat flux distributions at the glass surfaces were to a certain extent mirror images of the gas temperature distribution within the 02-flames. Considerably flatter net heat flux distributions are predicted, when the glass near the surface responds to the imprint of the net heat flux distributions through change of its temperature. This effect is modelled by the simple model for variable glass surface temperatures. Results obtained from this model are shown in Fig. 18a for air combustion and in Fig. 18b for the baseline 02 burner configuration. In the case of air combustion and variable glass temperatures (Case 1.1), the higher and lowest heat fluxes are only 83 kW/m2 and 47 kw/m2 , respectively. Similarly, for the baseline 02 Case with variable glass surface temperatures (Case 3.1l, highest and lowest heat fluxes amount to 80 kw/m2 and 50 kw/m~, respectively. Reslllting distributions of calculated effective surface temper . ~'· ~; r. ~s for air Case 1.1 and baseline 02 Case 3.1 are compared in Figs. 19a and 19b., respectively. }leak glass temperatures Ts gl max for air combustion reach 26630 F (1735 K) and the differences between this peak and the lowest temperatures predicted is dTs ,gl = 105 R (59 K). For 02 combustion, these numbers. are Ts ,gl,max = 26560 F (1731 K) and dTs ,gl = 90 R (50 K), respect1vely. As it can be expected, in all cases investigated with variable glass surface temperatures, maximum refractory temperatures increased by about 10 K (20F) compared to corresponding cases assuming a uniform effective glass surface temperature. The higher peak refractory temperatures are offset by lower refractory temperatures in furnace corners, especially in the 02 cases which exhibit a somewhat more inhomogeneous refractory temperature distribution. This can be seen in Figs. 20 and 21 which display the refractory temperature distribution predicted for the air -Case 1. 1 and for the 02 Case 3 . 1, respectively. 15 |