Search Results

A numerical study has been performed to investigate the soot emission from a high-speed single-cylinder direct injection diesel engine. It was shown that the current KIVA CFD code with the standard evaporation model could predict the experimental trend, where at a low speed running condition a higher smoke reading is reached when increasing the injector protrusion into the piston chamber and conversely a lower smoke reading was recorded for the same change in injector protrusion at a high running speed condition. Evidence of inappropriate air/fuel mixing was seen via rates of heat release analyses, especially in the high-speed conditions. Efforts to reduce this discrepancy by way of improvements to the KIVA breakup and evaporation models were made. Results of the modified models showed improvements in the vapor dispersion of the atomizing liquid jet, thus affecting the mixing rates and predicted smoke emissions.

For many years, compression ignition combustion has been studied by a combination of generic studies on fuel spray formation and analysis of results from single and multicylinder engines. The results and insight have been applied to design and develop advanced fuel injection equipment for high-speed direct injection engines. Experimental fuel injection equipments, including early common rail designs, have been matched to combustion chambers in single cylinder research engines to tackle the conflicting requirements of efficiency and minimum nitric oxide formation, combustion noise and soot. A clear strategy evolved from the work with experimental equipment that is being applied to multicylinder engines. If sufficient oxygen is available in the gas charge trapped in each cylinder, the LDCR common rail injection system will provide the fuel required to develop high torque at low engine speeds.

An experimental study has been carried out on a single-cylinder pressure-charged engine with a near quiescent combustion system. An improvement in the NOx/BSFC trade-off was achieved by two different approaches, namely exhaust gas recirculation and pilot injection. Without EGR, a reference EUI-200 system with 1900 bar peak injection pressure gave a low soot particulate level of 0.013 g/bhp h over a simulation of the US FTP cycle. The results with EGR show how higher levels of EGR can be used at more advanced injection timings to give substantially improved NOx versus BSFC results compared with timing retard alone. It was possible to reduce the NOx from 4.85 to 3.6 g/bhp h for no increase in BSFC over the simulated US FTP cycle and with a total calculated particulate of 0.075 g/bhph. The results with electronically-controlled pilot injection show improvements in NOx versus BSFC, lower NOx before HC increase with retard, or reduced combustion noise at certain test modes.

Fuels of widely varying properties are studied by injection of a single and well defined spray into an experimental diesel engine. Three optical techniques were developed to visualise liquid fuel, fuel vapour, flame, soot and individual droplets and their associated vapour trails. Liquid core length measurements are presented for diesel fuel, toluene, ethanol and sunflower oil. Computer model predictions show that an increase of the fuel mid-boiling point by 40°C gives a similar effect on liquid core length to an increase of 0.03mm in the nozzle hole diameter.

Emissions results are presented from an experimental study of an indirect injection diesel engine. The mass measurements of particulate emission are correlated with the measurements of smoke and HC emission. The correlation provides a less expensive way of carrying out preliminary combustion optimisation work. It is concluded that practically all of the particulate mass emission is accounted for by: (a) the black smoke or soot formed in the high temperature fuel-rich regions of the diffusion phase of burning, (b) that fraction (about 50%) of the total HC mass emission which condenses at the particulate sampling filter. The total HC mass emission is itself a function of three distinct sources in the combustion process. The understanding gained is then used to define three combustion ideals for optimising diesel combustion to minimise fuel consumption and emissions of smoke, NOX, particulates, HC, CO, odor and noise.

An analytical technique has been optimised for the measurement of the concentrations of diesel exhaust odorants. Application of this technique to combustion bomb studies shows that preflame reactions with diesel fuel produce high concentrations of odorants. The effects of engine variables on exhaust odorant concentrations are presented for direct and indirect injection engines. Analysis of these data shows that diesel exhaust odorants are produced from three sources: (a) the fuel-lean mixture produced during the ignition delay period, (b) fuel emptying from the nozzle sac volume of direct injection engines after injection, (c) a fuel-rich source which becomes significant at high load. The practical measures for control of odorants are outlined.

Experimental data are presented to show how diesel combustion systems respond to increase of fuel injection rate. Concepts of a fuel spray entrainment parameter, a maximum useful injection rate, and a condition termed ‘impingement’ are introduced to correlate and interpret widely differing responses. Best possible smoke and BSFC values in swirl type direct injection engines are obtained for injection rates 15% to 33% higher than normal values, but in practice lower rates must be used to satisfy emissions and other requirements. Engines with a high swirl rate and impingement give a superior ‘retardability’ for normal injection rates. Computer model calculations also show that there is a maximum useful injection rate and explain the relative fuel economies for different diesel combustion systems.

The paper considers the combustion regimes which lead to the emission of HC and CO in diesel combustion systems. Experimental data are presented to show the effect of a wide range of operating variables on the emission of HC and CO from a range of production automotive diesel engines. Analysis of the data shows that three localised sources account for most of the HC and CO emissions. The measures required to exploit the very low emissions potential of the diesel engine are discussed.

Experimental data on the concentration of hydrocarbons (HC) emitted in the exhaust are presented for both direct injection (DI) and indirect injection (IDI) engines and cover the effect of a wide range of engine operating parameters. The analysis shows that there are two main sources of HC in DI engines. One is due to the volume of fuel in the sac and holes of the injection nozzle and the other is from fuel premixed to leaner than lean limit conditions. Reduction of sac volume and reduction of ignition delay are effective in reducing the HC from the sac volume and lean limit sources respectively. Developments in fuel injection equipment and engine design to reduce HC emissions are outlined.

This paper consists of two parts. Part I concerns the effects of injection timing, injection rate, and air swirl on emission of smoke and gaseous pollutants from direct-injection diesel engines. Studies show that fuel-injection equipment and variables such as nozzle configuration affect pollutant production and emission because they affect fuel-air mixing. An increased rate of injection or air swirl increases the rate of fuel-air mixing and reduces the amount of exhaust smoke and its dependence on injection timing. An increase in rate or swirl ratio increases nitric oxide emission at a given injection timing, but the increase is relatively small compared with reduction obtained by retarding injection timing. Substantial retard, in conjunction with increased rate of fuel-air mixing, limits loss in engine efficiency. Part II reports development of a model for calculating soot and nitric oxide formation.