OCR Text |
Show Experiments As shown in Fig. 1, a cylindrical furnace like that usually employed for heating and thermally treating metals, was used in these experiments. Its standard thermal input was 200 Mcal/h. Burners fitted around the furnace heat the empty inner case, which usually contains material requiring treatment. The standard temperature was measured at T in the inner case. Table 1 shows the layout of the burners and chimney, as well as the burning conditions. Three types of high-velocity burner developed by Osaka Gas were used, as depicted in Fig. 2. The thermal input of each burner was 100 Mcal/h. The MK burner had the highest flame velocity and a constantly burning pilot burner. LC-h and LC-m were made by closing the end of the straight combustion pipe of LC burner, which has no pilot burner and a directly ignited main gas supply. Low-NOx type and general burners were not used in this test because our experiments and past experience have shown that high levels of NOx are generated when the furnace wall is hit by their flames, which often exceed 700 mm in length. Other types, such as catalytic combustion burners, are too costly to incorporate and create uneven temperature distribution in the furnace. Before starting the experiments, a combustion test was carried out in a furnace with internal dimensions of 1580x720x720 mm, forming a 0.82 m3 firing space, as shown in Fig. 3. The results plotted in Fig. 4 indicate that raising the flow speed of the burner lowers the concentration of NOx that it generates. The NOx concentration of MK high-velocity burner is only slightly higher than that of low-NOx burners. High-velocity burners which lack special NOx reduction measures presumably reduce NOx levels by the in-furnace exhaust gas recirculation (EGR) effect, brought about by the high speed of the combustion flame. The rate of NOx concentration increase with furnace temperature was almost the same in each case. 1) NOx under basic conditions Test results obtained under the conditions listed in Table 1 are plotted in Fig. 5. A comparison of Figs. 5 and 4 shows that the NOx concentrations for the LC-m burner were almost constant in each case. However, the NOx concentrations of the LC-h and MK types were slightly lower in Fig. 5. In Fig. 5 the thermal input was double and the firing space was half (0.40 m3 ) those in Fig. 4. Thus, it is thought that the increased flowing speed of the flue gas into the furnace enhanced the EGR effect. In the LC-m burner, the furnace wall closest to the flame top was at a high temperature (Fig. 6, NO.6); this increased thermal NOx generation and appeared to largely cancel the EGR effect. - 2 - |