Search Results

This paper presents the results of investigations that were carried out on a single-cylinder spray-guided direct injection engine. The effects of injection pressures of up to 1000 bar on combustion and emissions at the stability limit of stratified load are presented. It is known that at low engine speeds, problems in mixture preparation occur due to insufficient in-cylinder motion at higher loads in stratified charge operation. Additionally, adverse effects appear at high engine speeds due to limited time for vaporization and mixture formation. Therefore, investigations at various engine speeds ranging from 2000 to 4000 rpm were performed. As a baseline, a production multihole injector is compared with an injector that has been specially adapted for higher injection pressures.

In this work, the potential of laser-induced ignition to improve combustion initiation and heat release in a direct-injection engine is investigated by a combined experimental and numerical investigation. Laser ignition is studied in fuel/air mixtures with homogeneous equivalence ratio fields. The results provide knowledge about minimum ignition energies and the ignition limits of laser-induced ignition. Furthermore, in mixtures with nominally identical conditions, statistical variations of the ignition success are observed experimentally. These variations can be explained, based on numerical simulations, by fluctuations in the strain rate in the turbulent in-cylinder flow. Additionally, laser ignition in engines with a spray-guided combustion mode, with strongly inhomogeneous fuel/air mixtures, was investigated.

An important source for soot formation during the combustion of diesel engines with direct injection is the interaction of liquid fuel or a very rich air/fuel-mixture with the flame. This effect appears especially in modern direct injection engines where the injection is often split in a pre- and a main injection due to noise reasons. After the ignition of the pre-injected fuel a part of the main injection can interact with the flame still in liquid phase as the fuel is injected straight towards the already burning cylinder areas. This leads to high amounts of soot. The injection strategy for this experimental study overcomes this problem by separating the injections spatially and therefore on the one hand reduces the soot formation during the early stages of the combustion and on the other hand increases the soot oxidation later during the combustion. In particular an injection configuration is used which gives the degree of freedom to modify the injection in the described manner.

Given the fact that, in an endeavor to achieve the goals of engineering for a trade-off between cleaning up exhaust emissions and maximizing fuel economy, two main paths are being followed in advancing and optimizing SI-engine operating strategy in the upper part-load range. On the one hand, homogenization and operation in the compression ignition mode seem to offer a promising means of minimizing NOx emission by keeping the combustion temperature below the formation borderline and accepting a high cylinder-pressure gradient to obtain benefits in fuel economy. On the other hand, there are ambitions to widen the range of stratified operation using a supercharger or turbocharger. This way, efficiency of the engine cycle can be improved by operating at a higher global air-fuel ratio and, with this, a higher polytropic exponent, thereby taking the efficiency chain to a higher level.

Understanding the processes regarding fuel injection, vaporization and combustion during cold start is very important in order to reduce the HC-emission of gasoline engines. To learn more about the cold engine start-up process an experimental study on a 4.2 liter eight cylinder engine with gasoline direct injection was carried out. Parameters such as injection and ignition timing as well as the injection quantity were varied to get information about their effect on the combustion process and speed rise. Especially during engine run-up it is important to investigate every subsequent combustion. Therefore the engine was equipped with high pressure indication in each cylinder. The transient pressures and the instantaneous crankshaft speed of the engine were recorded by means of an indication system. Additionally a fast response flame ionization detector (FRFID) was applied to measure the transient HC-emissions during the first cycles of the engine.

A general definition and an index for the assessment of different engine knock behavior have been developed. The knock onset locations have been determined by piezoresistive pressure actuators and optical fiber probes in full load engine operation mode. The thermodynamic conditions at the knock onset locations have been quantified by CFD-calculations. Therefore the local fuel concentration, mixture temperature and residual gas concentration have been considered. These calculated thermodynamic conditions were further used to calculate the necessary volume of an exothermal center for the generation of the maximal measured pressure amplitudes.

The main objectives of this investigation are the visualisation of the flame propagation depending on different boundary conditions. Turbulence intensities, wall-temperatures and an air-fuel ratios were varied in a wide range. For the experiments a rapid compression machine with a quadratic piston and a good optical access is used which allows to observe the entire burning. With an additionally integrated turbulence generator it is possible to create a defined turbulence field in the burning chamber.

The mixture formation of a direct injection HCCI engine was optimized by a new nozzle geometry in combination with a multi-pulse injection scheme. To achieve this injection strategy, a highly flexible Piezo-Common-Rail injection system was used at the single cylinder research engine (DC BR 500, compression ratio reduced to 14:1). Optical probes for local, time resolved temperature and soot concentration measurement (Two-Color-Method) were adapted. Furthermore, a camera system was applied to detect the radiation of UV light (especially emitted by OH radicals which is an important indicator of the ignition process). So, the behaviors of different fuels in regard to the combustion process have been investigated. Diesel, SMDS and mixtures of n-heptane and iso-octane were tested with various EGR ratios, boost pressures and air/fuel ratios.

To operate gasoline direct injection engines at part load and in stratified mode the mixture formation has to fulfil several requirements. The complexity of this process requires - regarding a suitable mixture transportation and vaporisation of the fuel - an adjusted design of the combustion chamber and the intake ports to reliably place an ignitable mixture at ignition timing near the spark plug at any speed and load. Due to the inhomogeneous mixture distribution during stratified operation, the first combustion period is very sensitive to cycle-to-cycle variations. A reproducible mixture movement with high kinetic energy is necessary for stable engine operation with low fluctuations in the combustion process. Because of the high relevance of these facts, the effects of an adjustable air guiding system in the inlet manifold on in-cylinder flow, ignition and combustion using optical measurement techniques were investigated.

The objective of this study is to achieve a general understanding of gas exchange phenomena to develop a model for predicting the residual gas content. The knowledge of the cylinder-charge composition is important for the thermodynamic analysis of the combustion process of IC engines. Therefore, the amount of fresh air and fuel as well as the residual gas fraction has to be known. The residual gas mass strongly depends on valve train parameters and operating conditions. In this study, the residual gas fraction has been determined by using in-cylinder gas sampling from the combustion chamber of a 4-stroke SI engine. The gas sampling valve was flush-mounted to the combustion chamber walls. The gas samples were taken after the gas exchange and analysed for its CO2 concentration. In combination with the analysis of the exhaust gas composition, the calculation of the residual gas fraction is possible.

The primary objective of this paper is to compare different methods of knock detection based on cylinder pressure data, with special regard to identifying knocking cycles and detecting the onset of knock. These investigations have resulted in the development of a new knock detection method based on high-pass filtered heat release. Different signal characteristics have been considered. The model has been developed on the basis of experimental data for a four-valve production engine, and verified over a wide range of operating conditions. For the purpose of thermodynamic investigations, the new knock detection algorithm allows the determination of the engine operating points that correspond to the knock limit, and their mean crank angle of knock onset. The thermodynamic properties of the end gas at knock onset have been discussed using a zero-dimensional two-zone model.

To meet future emission standards with gasoline direct injection engines it is important to have a reliable process robustness during stratified charge operation. Especially engines with a wide spacing arrangement of fuel injector and spark plug which operate with an air-guided concept are very sensitive concerning misfire operation caused by cyclic variations of the mixture formation and transport. Primarily the turbulent in-cylinder gas motion and the interaction with the fuel injection indicate these fluctuations. To reduce these cycle-to-cycle variations and to generate a steady flow behavior an adjustable air-guiding system was developed and attached to the inlet port of a single-cylinder DI engine. The following examinations show that the air-guiding system can lead to a significant reduction of the cycle-to-cycle-variation of the in-cylinder air flow. As a result of these improvements, the deviation of imep in the fired engine decreases obviously.

This paper analyses the scavenge process of a conventional two-stroke engine in order to find ways to significantly reduce the scavenge losses by applying a combination of 1D and 3D simulation procedures. A special evaluation method was developed which allows a clear distinction between the main hydrocarbon loss mechanisms. Furthermore, the paper presents an approach to simulate the highly turbulent combustion at a speed of 9000 rpm. The results of the numerical investigations are compared with experimental results. The engine chosen for this purpose was a 64 cm3 four-port production two-stroke engine. The CFD calculations were performed using the finite volume CFD code STAR-CD. The mesh generation process was automated using pro*am. Combustion was modelled with the one-equation Weller flamelet model. The results of the present study show that the combination of 1D and 3D simulation procedures is a powerful tool for further investigations (e.g. stratified charge, GDI).

Determining locations at which knocking events appear frequently is a very important tool for optimizing the efficiency of S.I.-Engines. The knock behavior of two Direct-Injection Engines (one with wall guided fuel delivery - Mitsubishi GDI-engine - and a single-cylinder test engine with narrow arrangement of spark plug and injector) was tested and the location of appearing knocking was detected by the use of optical fiber technique. The optical investigations were conducted using a spark plug equipped with 8 optical sensors evenly distributed in a ring on the ground electrode of the standard spark plug. The intensity of flame radiation in the observing area of the optical sensors were measured. As knocking combustion produces pressure waves exciting gas oscillations, each compression of residual exhaust gas causes a steep increase of the measured flame intensity curve.

Reduction of fuel economy and exhaust emission at spark ignition engines with direct injection can be achieved by investigation and optimization of mixture preparation and combustion process. In this paper principle strategies of mixture preparation and combustion are discussed. Phenomena of combustion like flame radiation, flame propagation and knocking combustion are represented for different mixture preparation strategies. For detection of combustion phenomena optical fiber technique as well as new visualization device with an endoscope have been used. From the view of present knowledge, obtained with investigation of spark ignition engines with direct injection, it is an important target for future development of GDI engine technology to force activities of combustion process investigation at different mixture preparation strategies.

In this paper optical investigations of a gasoline direct injection engine with narrow spacing arrangement of spark plug and injector are presented. For the combustion analysis spectroscopy techniques based on the fiber technique are used. With this measurement technique information about soot formation and temperature progression in the combustion chamber is obtained. Furthermore a validation of numerical simulation of the stratified combustion with data obtained experimentally, is performed and discussed.

In this paper an estimation of efficiency potential of the engine process with Gasoline Direct Injection (GDI) is presented as well as both the advantages and todays problems of different mixture preparation concepts for the GDI engine. Furthermore examples of combustion analysis with optical measurement methods like Particle Image-Velocimetry (PIV) and spectroscopy techniques, which are important for future development steps in GDI, are shown and discussed. A validation of the numerical simulation of the stratified combustion process with data, obtained experimentally from a GDI engine, is performed and discussed. Consequently the combination of experimental and numerical methods provides both a better understanding of mixture preparation and combustion processes in GDI engines as well as an efficient development procedure for an optimized mixing and combustion process for future GDI engines.

The quality of air-/fuel-mixture is of prime importance for cycle fluctuations of combustion. Investigations of mixture formation and conditions in SI engines have been subject of intensive research since many years. The scope of this work was to investigate crank angle resolved determination of qualitative and quantitative mixture conditions inside the combustion chamber in dependence on various engine operating conditions. For this experimental investigation a time resolved Gas Sampling Valve (GSV) was combined with a flame ionisation detector (FID), a CO2-analyzer and a mass spectrometer. The GSV also enables the determination of residual gas concentration. Measurements on a DI gasoline engine show influences of air-/fuel-mixture in dependence on various engine operating conditions when the engine runs in charge stratification mode. Moreover, experimental results of local mixture composi-tion are compared with fuel distribution, calculated from CFD-codes.

The loss of momentum of the gas-core inside inlet manifolds of four-stroke engines is characterized by loss coefficients. Usually these coefficients are obtained by experimental investigations of the flow through cylinder heads under steady-state conditions. The dynamic behavior of the gas motion under real conditions due to acceleration and vibration of the gas-core as well as the influence of the gas motion due to the exhaust can not be described by these coefficients. Therefore a basic investigation of the unsteady flow under real engine conditions has been performed. The aim was to develop a simple method to characterize the inlet flow behavior under real conditions and to define a dynamic loss coefficient. The mass flow rate was determined by time resolved pressure data inside the suction pipe and a simple numerical calculation method considering unsteady flow conditions. The verification of calculated flow velocities was performed by using Particle-Image-Velocimetry.

A new fiber optic sensor has been used to detect knocking combustion. With this sensor it is possible to detect high frequency signals which are free from electrical and mechanical disturbance. By using the maximum signal rise of the detected optical signals for each combustion cycle, it is possible to clearly seperate knocking and non-knocking cycles. The detected maximum signal rise was used in a preliminary test as the input of a knock control system.