Impact of Secondary Air Flow on Combustor Emissions

Abstract

Abstract It is a common practice to relate emissions performance of Dry Low Emissions (DLE) combustion systems to the flame temperature that is estimated from the mass flows of air and fuel flowing through the premixer. In many combustion systems, the exit temperature (or turbine nozzle inlet temperature) is quite low and is not a good parameter for estimating combustion emissions. The difference between the combustion flame temperature and exit temperature is mainly due to secondary air dilution. To our knowledge there are no detailed published data that quantify the impact of this temperature difference on combustion emissions. The target of this study is to quantify the impact of secondary air variation on emissions, both globally and locally. High pressure experiments are conducted at H class gas turbine operating conditions using a DLE combustion system. In the context of this DLE system, secondary air refers to cooling and leakage flows because direct air dilution of the combustion gasses is not necessary. This is because the flame stabilized downstream of the premixer is well mixed and fuel-lean. With NOx requirements moving toward single digit (ppm) levels, it becomes essential to accurately quantify the impact of reducing the secondary air percentage on emissions performance. In addition to the need to carefully study the impact of local interaction of the secondary air with the flame. The combustion system is configured with two independently controlled mixers along with a variable secondary air circuit that can change the secondary air fraction from 14 to 8%. Multiple emissions rakes are used at the combustor exit to delineate the interaction and relate it to the flame structure. The system is configured to enable sampling from individual rakes to study local emissions and the rakes can be ganged together to measure the bulk-averaged combustion emissions. This research provides a quantification of the improvement of the NOx margin with a decrease in the secondary air percentage. The study shows that the increase in margin is not a simple re-estimate of the combustor emissions using the NOx design curve due to flame quenching effects. The results also show that the secondary air can be used to improve the NOx emissions via controlling the interaction with the primary flame. The impact is quantified in terms of emissions, acoustics and metal temperatures.