Book Description
The compression ignition (diesel) engine continues to be an attractive option for vehicle powertrains, particularly in heavy- and medium-duty transportation. While offering many operational advantages including a high power output per unit weight, quick refueling capabilities, and reliability; the diesel engine is prone to produce significant emissions. Chief among these emissions are soot and nitrogen oxides (NOx). Combustion techniques that reduce soot tend to increase NOx and vice versa requiring an acceptable soot-NOx tradeoff. The optimization of this tradeoff has been a challenge in recent decades with the advent of advanced combustion strategies and alternative fuels. Soot measurements at engine relevant conditions are the focus of this work and critical to building understanding of these complex phenomena and to evaluate potential solutions. Two-color pyrometry (2CP) has been used over several decades to study engine-relevant combustion processes, but results are generally regarded as qualitative or semi-quantitative. The objective of this work is to advance the 2CP diagnostic to achieve reliable, spatially-resolved measurements of the soot in diesel-like fuel sprays. This outcome is accomplished by 1) developing a new optical design for 2CP to overcome common measurement errors in other designs, 2) performing a detailed uncertainty analysis to quantify errors, and 3) applying the new 2CP system to statistically analyze the soot behavior in a diesel-like fuel spray. First, a novel optical configuration was developed and constructed from off-the-shelf components to eliminate systematic errors of previous designs. In many current 2CP systems, large measurement errors can be introduced by parallax because lines of sight (LsOS) of the two wavelengths are not the same. The modified optical hardware, was shown to accurately resolve corresponding pixels at both wavelengths of a high-resolution optical target, indicating that the LsOS for both wavelengths are the same. Next, an experimental investigation of reacting diesel-like fuel sprays revealed steady experiment conditions with repeatable global parameters of combustion. Despite the similarity of global behavior across injections, significant spatial variation of soot was observed between repeated injection experiments. Computed average turbulent flame speed for each injection suggests that initial reaction rates, which are a function of local equivalence ratio, determine the variation. Motivated by the spatial variations of soot observed across injections, the 2CP technique was further refined for quantitative, spatially resolved measurements. A pixel-by-pixel calibration was applied to account for any non-uniformities in sensor response, increasing accuracy of the computed soot quantities at each pixel. Uncertainty analysis determined the reliability of each individual measurement. This study found that the largest relative uncertainties are associated with low soot concentrations. Highly uncertain soot measurements typically occur on the edges of the diffusion flame where is soot is most likely to be oxidized. However, these highly uncertain data had marginal impact on the total soot mass produced in the flame. Finally, a large 500-injection data set employed all the previously developed 2CP diagnostic capabilities to evaluate spatiotemporal sooting behavior in a diesel-like spray flame. Focusing on the gaps in the literature for temporal development of soot and detailed study of the statistical variability between injections, this work provides insights on soot formation in key regions: lift-off, core, and jet head. Total soot contribution from binned individual soot mass values indicate that large soot masses take longer to form, but are the first to oxidize. Overall, the majority of injections produced low soot, but a few produced exceptionally high soot by comparison.