OCR Text |
Show As shown in Figure 5, the heat transfer mechanism includes radiative, convective and conductive heat transfer among combustion gases, the kiln wall and the solid bed. A two-sink model is used to calculate the radiative heat transfer within the freebo ard . The radiative heat transfer between the solid bed and the covered inne r wall is modeled as radiation between two infinite parallel plates(9]. Convective heat transfer coefficients at the exposed and covered areas are adopted from Gorog, et al(10] . The solid bed is considered to be well-mixed at any cross section, which is a reasonable assumption for this operation[ll]. The combustion gas temperature is also considered uniform, which is a good approximation for the flame characteristics of the "A" Burner (5&8]. The gas is taken to be radiatively gray in the calculation of gas emissivity. The model can be used to predict the fuel requirement, flue gas properties, and axial temperature profiles of the solid bed, the kiln inner wall and the outer skin. Figure 6 shows temperature profile predictions which indicates that heat transfer is not a limitation for the MIS at a 4000 lb/hr throughput . Projected Perfo~ance Appropriate heat and mass balances for the EPA/MIS were calculated before the Denney Farm tests using the aforementioned computer model(S]. The air case with a throughput of 2000 lb/hr of contaminated solid with 20~ moisture was chosen as the basis for comparison . A comparable calculation for an oxygen-fired rotary kiln was also made to predict the impact of using the Linde OCS. This comparison, summarized in Table 2, projected the following benefits using oxygen: (1) a throughput increase of 100~, i.e., a processing rate of 4,000 lb/hr is feasible . (2) such a throughput increase achievable with the oxygen system can reduce the incineration cost per ton of soil . Comparison of Calculated and Actual Results In the field demonstration of the EPA MIS, the air leakage rate into the rotary kiln was estimated to be about 8,000 SCFH at a vacuum level of 0 .2 to 0.3 inches of water column. The air leaked partially through the kiln seals between the stationary and rotary components and partially through the ram feed chute. In addition, 4,000 CFH of compressed air was used for the waste and water atomization nozzles. The backup air burner also needed 1,000 CFH of air flow to avoid overheating. The oxygen burner required only 400 SCFH of atomizing air . Altogether there was about 13,000 SCFH of air entering the kiln . The actual oxygen usage was lower than originally expected mainly due to the higher than expected air infiltration into the kiln. Therefore, only 58~ of the oxygen entering the kiln is supplied by pure oxygen (equivalent to a 40~ oxygen enrichment level), instead of the planned 95~ . There are other factors affecting the oxygen and fuel consumptions. The rotary kiln was operated at a lower temperature (around 1550°F) than the originally anticipated 1800°F . Also , the heat loss from the secondary combustion chamber of about one million BTU/HR was under-estimated by 40~ in the original estimation. Furthermore, it was - 7 - |