Search Results

This study investigates if particulates from a GDI engine can be significantly suppressed by use of ultra-high injection pressures under 2 different engine conditions known to be associated with high particulate numbers (PN): warm-up and transients. Experiments were carried out in a single-cylinder GDI engine equipped with an endoscope connected to a high-speed camera to enable combustion visualization. To mimic the warming-up, the coolant temperature was varied between 20 °C and 90 °C. A Diesel injector with modified nozzle was used and the injection pressures were varied between 400 and 1500 bar. The results revealed that increasing the fuel injection pressure decreased engine out HC and PN under warming-up conditions. However, the coolant water temperature was the most dominant factor affecting the emissions. For coolant temperature of 20 °C, the use of 1500 bar fuel injection pressure in comparison to lower fuel pressures resulted in significantly lower PN.

Homogeneous lean spark-ignited combustion is known for its thermodynamic advantages over conventional stoichiometric combustion but remains a challenge due to combustion instability, engine knock and NOx emissions especially at higher engine loads above the naturally aspirated limit. Investigations have shown that lean combustion can partly suppress knock, which is why the concept may be particularly advantageous in high load, boosted operation in downsized engines with high compression ratios. However, the authors have previously shown that this is not true for all cases due to the appearance of a lean load limit, which is defined by the convergence of the knock limit and combustion stability limit. Therefore, further research has been conducted with the alternative and potentially renewable fuel methane which has higher resistance to autoignition compared to gasoline.

To assist efforts reducing harmful emissions from internal combustion engines, particulate formation was investigated in a compressed natural gas (CNG) Direct Injection single-cylinder SI engine in warm-up conditions. This involved tests at low engine speed and load, with selected engine coolant temperatures ranging from 15 to 90 °C, and use of a gasoline direct injection (GDI) system as a standard reference system. Total particulate number (PN), their size distribution, standard emissions, fuel consumption and rate of heat release were analyzed, and an endoscope with high-speed video imaging was used to observe combustion luminescence and soot formation-related phenomena. The results show that PN was strongly influenced by changes in coolant water temperature in both the CNG DI and GDI systems. However, the CNG DI engine generated 1 to 2 orders of magnitude lower PN than the GDI system at all tested temperatures.

To characterize the effects of renewable fuels on particulate emissions from GDI engines, engine experiments were conducted using EN228-compliant gasoline fuel blends containing no oxygenates, 10% ethanol (EtOH), or 22% ethyl tert-butyl ether (ETBE). The experiments were conducted in a single cylinder GDI engine using a 6-hole fuel injector operated at 200 bar injection pressure. Both PN in raw exhaust and solid PN (SPN) were measured at two load points and various start of injection (SOI) timings. Raw PN and SPN results were classified into various size ranges, corresponding to current and future legislations. At early SOI timings, where particulate formation is dominated by diffusion flames on the piston due to liquid film, the oxygenated blends yielded dramatically higher PN and SPN emissions than reference gasoline because of fuel effects.

Public health risk and resulting stringent emission regulations for internal combustion engines pose a need for solutions to reduce particle emissions (PN). Current PN control approaches include increasing fuel injection pressure, optimizing spray targeting, multiple injection strategies, and the use of tumble flaps together with gasoline particulate filters (GPF). Experiments were performed using a single-cylinder spark-ignited GDI engine equipped with a custom inlet manifold and a port fuel injector located 500 mm upstream. Particulate emissions were measured during stationary medium/high load operation to evaluate the effect of varying the mass split between the direct and upstream injectors. Mixing quality is improved substantially by upstream injection and can thus be controlled by altering the mass split between the injectors.

48V mild hybrid powertrains are promising technologies for cost-effective compliance with future CO2 emissions standards. Current 48V powertrains with integrated belt starter generators (P0) with downsized engines achieve CO2 emissions of 95 g/km in the NEDC. However, to reach 75 g/km, it may be necessary to combine new 48V powertrain architectures with alternative fuels. Therefore, this paper compares CO2 emissions from different 48V powertrain architectures (P0, P1, P2, P3) with different electric power levels under various driving cycles (NEDC, WLTC, and RTS95). A numerical model of a compact class passenger car with a 48V powertrain was created and experimental fuel consumption maps for engines running on different fuels (gasoline, Diesel, E85, CNG) were used to simulate its CO2 emissions. The simulation results were analysed to determine why specific powertrain combinations were more efficient under certain driving conditions.

Maximum fuel injection pressures for GDI engines is expected to increase due to positive effects on emissions and engine-efficiency. Current GDI injectors have maximum operating pressures of 35 MPa, but higher injection pressures have yielded promising reductions in particle number (PN) and improved combustion stability. However, the mechanisms responsible for these effects are poorly understood, and there have been few studies on fuel sprays formed at high injection pressures. This paper summarizes experimental studies on the properties of sprays formed at high injection pressures. The results of these experiments can be used as inputs for CFD simulations and studies on combustion behavior, emissions formation, and combustion system design. The experiments were conducted using an injection rate meter and optical methods in a constant volume spray chamber. Injection rate measurements were performed to determine the injectors’ flow characteristics.

Natural gas, biogas, and biomethane are attractive fuels for compressed natural gas (CNG) engines because of their beneficial physical and chemical characteristics. This paper examines three combustion modes - homogeneous stoichiometric, homogeneous lean burn, and stratified combustion - in an optical single cylinder engine with a gas direct injection system operating with an injection pressure of 18 bar. The combustion process in each mode was characterized by indicated parameters, recording combustion images, and analysing combustion chemiluminescence emission spectra. Pure methane, which is the main component of CNG (up to 98%) or biomethane (> 98 %), was used as the fuel. Chemiluminescence emission spectrum analysis showed that OH* and CN* peaks appeared at their characteristic wavelengths in all three combustion modes. The peak of OH* and broadband CO2* intensities were strongly dependent on the air/fuel ratio conditions in the cylinder.

The current trend toward more fuel efficient vehicles with lower emission levels has prompted development of new combustion techniques for use in gasoline engines. Stratified combustion has been shown to be a promising approach for increasing the fuel efficiency. However, this technique is hampered by drawbacks such as increased particulate and standard emissions. This study attempts to address the issues of increased emission levels by investigating the influence of high frequency ionizing ignition systems, 350 bar fuel injection pressure and various tumble levels on particulate emissions and combustion characteristics in an optical SGDI engine operated in stratified mode on isooctane. Tests were performed at one engine load of 2.63 bar BMEP and speed of 1200 rpm. Combustion was recorded with two high speed color cameras from bottom and side views using optical filters for OH and soot luminescence.

Boosting and stratified operation can be used to increase the fuel efficiency of modern gasoline direct-injected (GDI) engines. In modern downsized GDI engines, boosting is standard to achieve a high power output. However, boosted GDI-engines have mostly been operated in homogenous mode and little is known about the effects of operating a boosted GDI-engine in stratified mode. This study employed optical and metal engines to examine how boosting influences combustion and particulate emission formation in a spray-guided GDI (SGDI), single cylinder research engine. The setup of the optical and metal engines was identical except the optical engine allowed optical access through the piston and cylinder liner. The engines were operated in steady state mode at five different engine operating points representing various loads and speeds. The engines were boosted with compressed air and operated at three levels of boost, as well as atmospheric pressure for comparison.

Among many techniques used for increasing fuel efficiency of a modern Gasoline Direct-Injected (GDI) engine are boosting and stratified operation. In modern downsized GDI engines, boosting is standard in order to achieve a high power output. Boosted GDI-engines have however mostly been operated in homogenous mode and little is known on the effects of operating a boosted GDI-engine in stratified mode. This paper presents the influence on combustion, standard emissions and particulate size distribution in a Spray-Guided, Gasoline, Direct-Injected (SGDI), single cylinder, research engine operated with various levels of boost. The engine was operated in steady state mode at five engine operating points of various load and speed. The engine was boosted with a Roots blower and operated at four levels of boost as well as atmospheric pressure for comparison. The engine was fueled with market gasoline (95 RON) blended with 10% ethanol.

Time-resolved mass injection rates of an outward opening piezo-actuated and a solenoid actuated multi-hole GDI injector were measured to investigate (1) the influence of both hardware and software settings and (2) the influence on the injection rates from a wide range of operational parameters and (3) discuss limitations and issues with this measurement technique. The varied operating parameters were fuel pressure, back-pressure, electrical pulse width, single/double injection and injection frequency. The varied hardware/software parameters were injector protrusion, upstream fuel pressure condition and the cut-off frequency of the software's low-pass filter. Signal quality was found to be dependent on both hardware and software settings, especially the cut-off frequency of the low-pass filter. Measurements with high signal quality were not possible for back-pressures lower than 0.5 MPa.

Transient compressible gas jets, as encountered in direct injection gas fuel engines, have been examined using Schlieren visualization. Helium has been injected into air in a pressure chamber to create the jets examined. The structure of the jets is studied from the mean and coefficient of variation of the penetration length, jet width and jet angle. The quantities are calculated by digital image processing of Schlieren images captured with a high-speed camera. Injection pressure and chamber pressure have been varied to determine whether they have an effect on the response variables. Design of experiments methods have been used to develop the scheme employed in performing the experiments. The mean normalized penetration length of the jets is found to scale with injection to chamber pressure ratio and is in agreement with a momentum conserving relation given in the literature. The dispersion of the penetration length has been found to be in agreement with a normal distribution.

The evaporation of different fuels and fuel components in hollow-cone sprays at conditions similar to those at stratified cold start has been investigated using a combination of planar laser-induced fluorescence (LIF) and Mie scattering. Ketones of different volatility were used as fluorescent tracers for different fuel components in gasoline-like model fuels and ethanol-based fuels. LIF and Mie images were compared to evaluate to what extent various fuel components had evaporated and obtained a spatial distribution different from that of the liquid drops, as a function of fuel temperature and time after start of injection. A selective and sequential evaporation of fuel components of different volatility was found.

Gasoline and gasoline-ethanol sprays from an outward-opening piezo-injector were studied in a constant volume/pressure chamber using high-speed imaging and phase doppler anemometry (PDA) under stratified cold start conditions corresponding to a vehicle ambient temperature of 243 K (−30°C/−22°F); in-cylinder air pressure of 5 bar, air temperature of 350 K (−30°C/−22°F) and fuel temperature of 243 K. The effects of varying in-cylinder pressure and temperature, fuel injection pressure and fuel temperature on the formation of gasoline, E75 and pure ethanol sprays were investigated. The results indicate that fuel composition affects spray behaviour, but less than expected. Furthermore, varying the temperature of the fuel or the air surrounding the spray also had minor effects. As expected, the fuel injection pressure was found to have the strongest influence on spray formation under stratified conditions.

The spray formation and consequent atomization of an outward opening piezo-electric gasoline DI injector have been experimentally investigated in a constant pressure spray chamber. The sizes and velocities of the droplets and the resulting spray shape were evaluated, under different boundary conditions, using Planar Mie scattering and Planar Laser-induced Fluorescence (PLIF) in combination with Phase Doppler Anemometry (PDA) analyses and high-speed video photography. The use of piezo-electric actuation for gasoline DI injectors provides an additional means to control the atomization and spray shape that is not available with solenoid-driven injectors such as swirling and multi-hole type injectors. For instance, with piezo injectors up to four injections per cycle are possible, and the fuel flow rate can be controlled by adjusting needle lift. The captured high-speed video images show that a hollow-cone spray forms as the fuel exits the outward-opening nozzle.