OCR Text |
Show Because of the measurement problems, many have concluded that direct control of fuel-air ratio is possible only on a limited basis, largely confined to small, single-burner boilers or process heaters. An often stated opinion maintains that only total average adjustments can be implemented on units with multiple burners; as one maladjusted mixer or burner will affect the overall results, with no means of determining which is the faulty unit. Based on experience with multiple burner equipment, we are convinced that flue gases do not mix thoroughly in exit ducts of fUrnaces. Large combustion equipment, such as utility steam generators, show definite streaming, or stratification of, exhaust gases. This presents the possibility of tracing excess oxygen or combustibles in exit duct areas to specific burners. Evidence of this relationship on multi-burner combustion equipment will provide the basis for improved efficiency in fUel utilization by better adjustment of fUel-air mixture at each burner. Additional benefits will be realized by improved burner balance and faster, positive, identification of malfUnctioning equipment. Sampling and analysis within gas streams flowing from burners reveal flow patterns easily detected at considerable distance from the burners. An obstruction, such as a boiler tube, will create some disturbance, but mixing is quite localized and has little effect on total flow pattern results. Observations have shown that stratified patterns originating from burners can still be identified after flowing by many such obstructions. Induced draft fans effectively mix the gases enough that the stratification is no longer identifiable. Leaks in equipment, such as air heaters, will cause contamination of samples, leading to errors in measurement. Analyzer sample points must, therefore, be located ahead of such devices. The nearer the exit from the combustion chamber, the more pronounced the stratification observed. This constitutes no problem for the fast-response, high temperature oxygen and combustibles analyzers available today. Thus, practical use can be made of multipoint analysis of the stratified gas streams and individual burners may be adjusted for maximum efficiency, either manually or automatically. The practical value of this technique lies in the potential for determining the combustion conditions for each burner on any multiburner apparatus. Maximum combustion efficiency is typically limited by any burner that happens to be set fUel-rich. Overall reduction in excess air is limited to the pOint where the richest burner generates significant amounts of carbon monoxide; thus the entire boiler or heater may well be operating with unnecessarily high excess air. By better balancing of the individual 230 burner fUel-air ratios, efficiency will be improved and fUel costs reduced. Our utility partner in the EPRI sponsored program has published findings on the stratification effects and uses gas analyzer data to monitor individual burner performance on multiple burner oil and gas fired boilers. We have much evidence, based on informal observations, that the stratified flue gas pattern can be traced to individual burners on many smaller industrial boilers and process heaters. Demonstration of such a correlation will lead to better combustion efficiency through improved adjustment of individual burners. CASE HISTORIES FIRST DOE INDUSTRIAL SITE DESCRIPTION, PROCEDURES, AND RESULTS - The boiler at the first DOE industrial test site was a water tubed, balanced draft type with a maximum capacity of approximately 90,000 Ibs. steam/hour. Four burners are arranged in a rectangular pattern on the boiler front. The usual fuel is pul veri zed coal, but oil could be burned as well. The hot combustion gas makes three passes through the tubes involving three 900 and two 1800 turns (all in the same plane) before entering the air heater at the boiler exit. There were no superheaters, and the internal walls and the baffles were in good condition. Our sample probes were located between the last hot gas bend just before the air heater as shown in Figure 1. Duct dimensions at the sample point were 3 ft. high by 17 ft. wide. Data were taken to map oxygen content (representing excess air present) at several locations within the duct. The data plots were used as basis for positioning the sample probe of a continuous analyzer for later tests directed toward confirming the stratification hypothesis. In a following series of tests, a sample probe was fixed in a selected location and connected to a continuous oxygen analyzer and recorder. Perturbations of the fUel-air ratio of each burner in turn were made by monitoring injection of oil, and the effect on oxygen level at the sample location was recorded. The probe was then repositioned to another area identified by the earlier traverses and the ratio variations repeated. This procedure continued until the effect of all four burners in four duct areas had been recorded. Figure 2 diagrams the relationship of burners to duct locations and also shows the configuration of the piping transporting pulverized coal from the mills to the burners. In Figure 3, the results of oxygen analyzer meas- |