| OCR Text |
Show AFRC - 1996 Int. Symposium Baltimore September 30 - October 2 1996 Page 2 of 14 expected. This obliges manufacturers to develop and optimize combustion equipment to achieve minimum NOx and CO levels. The NOx emission from a natural gas flame is primarily dependent on the peak flame temperature. Therefore, the key to decrease the NOx emission of natural gas flames is to lower this peak flame temperature. The use of relatively cool flames may, however, result in reduced flame stability and/or higher CO emission. This limits the potential using the peak flame temperature as a NOx control strategy. In practice, an optimum has to be found to achieve minimum NOx emission while maintaining good flame stability. Several techniques to reduce the flame temperature have been developed, such as air or fuel staging and internal or external flue-gas recirculation. A characteristic feature of each of these techniques is that combustion takes place in a mixture of gases containing fuel, air and flue-gas. A difference between these techniques is that in the case of external flue-gas recirculation, the flue-gases "added" to the flame are relatively cold (= 200°C), while in the case of internal flue gas recirculation, the flue gases have a temperature similar to the furnace temperature (e.g. 1000°C). When air or fuel staging is employed, the flue gases in the main flame originate usually both from the ftrst stage and from the furnace. The flue gases from the first stage can have temperatures close to the adiabatic flame temperature (1800-2000°C) while the flue gases originating from the furnace will have a temperature similar to the furnace temperature. This paper deals with research on the stability and NOx emission of flames fed by mixtures of flue-gas, air and natural gas. The temperatures of these mixtures have been varied to simulate the various options to add flue gases to the flame. Combustion experiments have been carried out in a relatively simple geometry. In addition, model calculations to' obtain the laminar burning velocities of the mixtures have been performed, and the results are compared with the experimental data gathered. The aim of the study is to obtain "engineering rules" with respect to flame stability and NOx formation as influenced by the composition of the initial (unburnt) mixture and by its initial temperature. 2. EXPERIMENT AL SET-UP Figure 1 shows a scheme of the experimental facility used. It consists of a precombustor, a venturi mixer and an air-cooled flame tube. The flame tube has a length of 1000 mm and an inner diameter of 180 mm. At the front plate of the flame tube, a metal-ftber flame holder is mounted to stabilize the combustion. The diameter of the flame holder is the same as the inner diameter of the flame tube (180 mm). This is to prevent the entrainment of extra fluegases from inside the flame tube into the flames under investigation. |
