OCR Text |
Show FLUE GAS STRATIFICATION IN INDUSTRIAL AND UTILITY BOILERS Marshall H. Cooper, Richard K. Jahn Bailey Controls Company Wickliffe, Ohio, USA PROGRAM DESCRIPTIONS Two independent projects, separately sponsored by the Department of Energy and the Electric Power Research Institute, were put in place to prove the existence of flue gas stratification in industrial and utility type bOilers. The objective of these projects was to demonstrate the feasibility of using the analysis of stratified flue gas as a guide to the more precise adjustment of burners, thereby increasing the fuel efficiency of mul tiple burner combustion equipment. Under the DOE sponsored project, our efforts were directed toward industrial boilers. We located two local industrial sites who were agreeable to participating with us under this program. The EPRI sponsored project was a joint effort between the Bailey Controls Company and a very willing utility partner. The utility believed in the stratification phenomenon and had in fact been using it for years to locate and adjust maladjusted burners. Numerous papers have been published which referenced their work in this area. We will also discuss flue gas stratification measured on a very small, two burner gas fired industrial boiler as well as on a larger three cyclone burner utility boiler. Before going into the test results, a brief review of combustion and stratification theory and their interaction is in order. THEORY OF COMBUSTION AND STRATIFICATION - The introductory chapters of reference handbooks on combustion make the process deceptively simple - and indeed, the chemical reactions are not complicated. Basically, a fuel containing hydrogen and carbon is brought into contact with air containing oxygen. Mixed in proper proportions, maintained at a temperature high enough to sustain the reaction, and given sufficient time, theoretically, perfect combustion will resul t. Unfortunately, in the real world, maintaining 229 these fundamental requirements - time, temperature, and turbulence for mixing - is not as easily achieved as theory might suggest. Specifically, ideal mixing of fuel and air is achievable only at great cost and can be easily affected by fuel or air variations, dirt, wear, and other factors. Maintaining proper fuel-air ratio is also not easy, especially when using fuels that may vary in composition. Time and temperature often interact. If the furnace volume is insufficient to provide time for the fuel to burn completely, the flame may be quenched before combustion is complete. Because of such conditions, it cannot be assumed that supplying a slight excess of air will lead to the complete combustion of the fuel. It is usual to find significant quantities of combustible material such as carbon monoxide in flue gases even though excess air is also present. A complete list of factors influencing the combustion process would not only be very long but would vary from one installation to another. One of the most troublesome problems in applying analyzers to monitor flue gases is to insure that the substance being analyzed is representative of the products of combustion leaving the furnace. It is often surmised that hot gases flowing through a duct diffuse rapidly into a homogeneous mixture. This theory is not supported by experience. Wide variations in composition exist from one place to another in a flue. Complicated, high-maintenance systems have been devised to combine samples from different points in order to produce a representative average. Another approach places the analyzer far downstream from the furnace exit, sometimes even beyond fans, filters, or scrubbers, on the assumption that the additional time and volume will allow more complete mixing. Such methods are self defeating, however, because the time lags in measurement and attendant loss of resolution detract from the usefulness of the signal. |