Search Results

A prior study (Chon and Heywood, [1]) examined how the design and performance of spark-ignition engines evolved in the United States during the 1980s and 1990s. This paper carries out a similar analysis of trends in basic engine design and performance characteristics over the past decade. Available databases on engine specifications in the U.S., Europe, and Japan were used as the sources of information. Parameters analyzed were maximum torque, power, and speed; number of cylinders and engine configuration, cylinder displacement, bore, stroke, compression ratio; valvetrain configuration, number of valves and their control; port or direct fuel injection; naturally-aspirated or turbocharged engine concepts; spark-ignition and diesel engines. Design features are correlated with these engine’s performance parameters, normalized by engine and cylinder displacement.

There is a trend towards increasing the degree of engine downsizing due to its potential for reducing fuel consumption and hence lowering CO2 emissions. However, downsizing introduces significant challenges for the engine airpath hardware and control, if driveability is to be maintained at an acceptable level. The transient response of the engine is affected by both the hardware selection and the associated controller. In order to understand the potential performance and limitations of the possible airpath hardware, a mean value model of the engine under consideration can be utilized. One benefit of these models is that they can be used as the basis of a model predictive controller which gives close to optimal performance with minimal tuning effort. In this paper we examine different two-stage series sequential turbocharger arrangements.

Understanding mixture formation phenomena during the first few cycles of an engine cold start is extremely important for achieving the minimum engine-out emission levels at the time when the catalytic converter is not yet operational. Of special importance is the structure of the charge (film, droplets and vapour) which enters the cylinder during this time interval as well as its concentration profile. However, direct experimental studies of the fuel behaviour in the inlet port have so far been less than fully successful due to the brevity of the process and lack of a suitable experimental technique. We present measurements of the hydrocarbon (HC) concentration in the manifold and port of a production SI engine using the Fast Response Flame Ionisation Detector (FRFID).

This paper describes the results of experimental and computer simulation studies of the combustion process in the prechamber three-valve stratified-charge engine. Prechamber and main-chamber pressure data and matched computer simulation calculations are used to determine the effects of variations in overall air/fuel ratio, engine speed and load, and prechamber volume and orifice diameter on the parameters which define the combustion process (spark advance for optimum torque, ignition delay, combustion duration), on cylinder pressure diagrams (mean main-chamber pressure, mean pressure difference across the orifice, and cycle-by-cycle pressure fluctuations) and on exhaust emissions. General correlations are derived from the data for the shape of the combustion rate profile and the extent of the combustion duration.

The effects of injection variability, low velocity fuel injection, and injector orifice size on unburned hydrocarbon emissions were studied in a direct-injection stratified-charge (DISC) engine. The engine incorporated a combustion process similar to the Texaco Controlled Combustion System (TCCS) and was operated with gasoline. The variability in the amount of fuel injected per cycle was found to have a negligible effect on HC emissions. Changing the amount of fuel injected at low velocity at the end of injection impacted the HC emissions by up to 50%. A positive pressure differential between the injection line and the combustion chamber when the injector needle closed resulted in more fuel injected at low velocity and increased HC emissions. High speed single frame photography was used to observe the end of injection. Injectors with smaller orifices had substantially lower HC emissions than the baseline injector.

The gas velocity through an automotive catalyst has been determined by measuring the time of flight of a pulse of propane injected at the inlet plane of the catalyst. The arrival time at the exit plane was detected by a fast flame ionization detector. By synchronizing and delaying the injection of propane with respect to the engine crankshaft position, the fluctuations of the exhaust gas velocity during the engine cycle were investigated. A number of tests at different engine load and speed points were carried out. The results show a complex velocity/time characteristic, including flow reversals. The technique is shown to be a viable option for flow measurement in this harsh environment.

To understand the effects of crevices on the engine-out hydrocarbon emissions, a series of engine experiments was carried out with different piston crevice volumes and with simulated head gasket crevices. The engine-out HC level was found to be modestly sensitive to the piston crevice size in both the warmed-up and the cold engines, but more sensitive to the crevice volume in the head gasket region. A substantial decrease in HC in the cold-to-warm-up engine transition was observed and is attributed mostly to the change in port oxidation.

Light-load unburned hydrocarbon emissions were studied experimentally in a spark-ignited direct-injection engine burning gasoline where the piston temperature was varied. The test engine was a single-cylinder Direct Injection Stratified-Charge (DISC) engine incorporating a combustion process similar to the Texaco Controlled Combustion System. At a single low load operating condition, the piston temperature was varied by 50 K by controlling the cooling water and oil temperature. The effect of this change on unburned hydrocarbon emissions and heat release profiles was studied. It was found that by carefully controlling the intake air temperature and pressure to maintain constant in-cylinder conditions at the time of injection, the change in piston temperature did not have a significant effect on the unburned hydrocarbon emissions from the engine.

The way In which combustion chamber geometry affects combustion in SI engines was studied using a quasi-diraensional cycle simulation. Calculations were performed to investigate the following questions: (i) the sensitivity of geometric effects on combustion to engine operating conditions; (ii) the differences in burn duration between ten chamber geometries and spark plug locations; and (iii) the relative merits of improved chamber design and amplified turbulence as means to reduce burn duration. The results from these studies are presented and discussed.

The amount of residual gas present in the cylinder has a well documented effect on the combustion event at idle. The high levels of burnt gas present at low engine speed leads to significant cyclic variability. This paper presents research which indicates that the temperature of the residual gas, which can vary from event to event depending on the spark timing, also has a significant effect on the combustion torque. The more the spark timing is retarded from MBT timing, the more thermal energy is present in the exhaust gas. The idle speed control strategy typically varies the spark to give fast torque actuation for good speed regulation and hence the temperature of the residual gas may change significantly within the space of a few events. The paper shows evidence of the phenomenon (with fixed engine speed and air mass flow) and discusses possible causes. It then proceeds to develop a dynamic model for the behaviour.

As a part of the effort to comply with increasingly stringent emission standards, engine manufacturers strive to minimize engine oil consumption. This requires the advancement of the understanding of the characteristics, sources, and driving mechanisms of oil consumption. This paper presents a combined theoretical and experimental approach to separate and quantify different oil consumption sources in a production spark ignition engine at different speed and load conditions. A sulfur tracer method was used to measure the dependence of oil consumption on engine operating speed and load. Liquid oil distribution on the piston was studied using a Laser-Induced-Fluorescence (LIF) technique. In addition, important in-cylinder parameters for oil transport and oil consumption, such as liner temperatures and land pressures, were measured.

A computer simulation of the four-stroke spark-ignition engine cycle has been used to examine the effects of turbocharging and reduced heat transfer on engine performance, efficiency and NOx emissions. The simulation computes the flows into and out of the engine, calculates the changes in thermodynamic properties and composition of the unburned and burned gas mixtures within the cylinder through the engine cycle due to work, heat and mass transfers, and follows the kinetics of NO formation and decomposition in the burned gas. The combustion process is specified as an input to the program through use of a normalized rate of mass burning profile. From this information, the simulation computes engine power, fuel consumption and NOx emissions. Wide-open-trottle predictions made with the simulation were compared with experimental data from a 5.7ℓ naturally-aspirated and a 3.8ℓ turbocharged production engine.

The design and operating characteristics of a single-cylinder transparent spark-ignition engine for Schlieren flow visualization are described. The engine is built on a CFR engine crankcase using the CFR piston and cylinder as a crosshead for the square cross-section piston and cylinder assembly. The square cross-section assembly has two parallel steel walls and two parallel quartz glass walls to permit optical access to the entire cylinder volume over the complete engine operating cycle. The CFR head and valve mechanism completes the assembly. It is shown that the engine operates satisfactorily with propane fuel under typical engine operating conditions. Schlieren short time-exposure photographs and high speed movies were taken to define details of the flow and density fields through the engine cycle. Photographs which illustrate key features of these fields are presented and described.

A new fast response NO detector, based on the chemiluminescence (CLD) method has been used to measure continuous, real time levels of NO in the cylinder, and simultaneously in the exhaust port of a virtually unmodified production SI engine. The real time NO concentration data show a great deal of information. Simultaneous NO measurements taken in-cylinder at sample points a few millimetres apart show substantial differences. Exhaust and in-cylinder levels from the same cycle show even greater differences, though the levels on average are well correlated.

The prediction of knock onset in spark-ignition engines requires a chemical model for the autoignition of the hydrocarbon fuel-air mixture, and a description of the unburned end-gas thermal state. Previous studies have shown that a reduced chemistry model developed by Keck et al. adequately predicts the initiation of autoignition. However, the combined effects of heat transfer and compression on the state of the end gas have not been thoroughly investigated. The importance of end-gas heat transfer was studied with the objective of improving the ability of our knock model to predict knock onset over a wide range of engine conditions. This was achieved through changing the thermal environment of the end gas by either varying the inlet air temperature or the coolant temperature. Results show that there is significant heating of the in-cylinder charge during intake and a substantial part of the compression process.

A gasoline direct injection fuel spray was observed using a fired, optical access, square cross-section single cylinder research engine and high-speed video imaging. Spray interaction with the piston is described qualitatively, and the results are compared with Computational Fluid Dynamics (CFD) simulation results using KIVA-3V version 2. CFD simulations predicted that within the operating window for stratified charge operation, between 1% and 4% of the injected fuel would remain on the piston as a liquid film, dependent primarily on piston temperature. The experimental results support the CFD simulations qualitatively, but the amount of fuel film remaining on the piston appears to be under-predicted. High-speed video footage shows a vigorous spray impingement on the piston crown, resulting in vapor production.

Photographic and performance studies with a Rapid Compression Machine at the Massachusetts Institute of Technology have been used to develop insight into the role of mixing in diesel engine combustion. Combustion photographs and performance data were analyzed. The experiments simulate a single fuel spray in an open chamber diesel engine with direct injection. The effects of droplet formation and evaporation on mixing are examined. It is concluded that mixing is controlled by the rate of entrainment of air by the fuel spray rather than the dynamics of single droplets. Experimental data on the geometry of a jet in a quiescent combustion chamber were compared with a two-phase jet model; a jet model based on empirical turbulent entrainment coefficients was developed to predict the motion of a fuel jet in a combustion chamber with swirl. Good agreement between theory and experiment was obtained.

Experiments were carried out to collect in-cylinder pressure data and microphone signals from a single-cylinder test engine using spark timingsbefore, at, and after knock onset for toluene reference fuels. The objective was to gain insight into the phenomena that determine knock onset, detected by an external microphone. In particular, the study examines how the end-gas autoignition process changes as the engine's spark timing is advanced through the borderline knock limit into the engine's knocking regime. Fast Fourier transforms (FFT) and bandpass filtering techniques were used to process the recorded cylinder pressure data to determine knock intensities for each cycle. Two characteristic pressure oscillation frequencies were detected: a peak just above 6 kHz and a range of peaks in the 15-22 kHz range. The microphone data shows that the audible knock signal has the same 6 kHz peak.

This study examines the scaling between engine performance, engine configuration, and engine size and geometry, for modern spark-ignition engines. It focuses especially on design features that impact engine breathing. We also analyze historical trends to illustrate how changes in technology have improved engine performance. Different geometric parameters such as cylinder displacement, piston area, number of cylinders, number of valves per cylinder, bore to stroke ratio, and compression ratio, in appropriate combinations, are correlated to engine performance parameters, namely maximum torque, power and brake mean effective pressure, to determine the relationships or scaling laws that best fit the data. Engine specifications from 1999 model year vehicles sold in the United States were compiled into a database and separated into two-, three-, and four-valves-per-cylinder engine categories.

The parameterization of variable geometry turbochargers for mean-value modeling is typically based on compressor and turbine flow and efficiency maps provided by the supplier. At low turbocharger speeds, and hence low airflows, the heat exchange via the turbocharger housing affects the temperature-based measurements of the efficiencies. Therefore, the low-speed operating regime of the turbocharger is excluded from the supplied maps and mean-value models mainly rely on extrapolation into this region, which is regularly met in emission drive cycles, and hence of significance. This paper presents experimental data from a 2.0-liter turbocharged common-rail diesel engine. While the flow maps extend from the high-speed region in a natural way, the efficiency maps are severely affected by the heat transfer effect. It is argued that this effect should be included in the mean-value model.