Search Results

The in-cylinder bulk flow and turbulence characteristics in a spark ignition engine were examined using two particle image velocimetry (PIV) systems. The effects of a turbulence-generating valve (TGV) bulk flow and turbulence characteristics were investigated. The results show that both the motion and location of turbulent flow can be controlled by a TGV valve. This could be used to reduce the level of unburned hydrocarbons produced during a cold start. The time history and scales of turbulence [1, 2 and 3] were compared with those obtained using laser Doppler velocimetry. The developed PIV systems were accurate, and the measured data were reliable enough to permit discussion of the cycle-resolve and cycle-to-cycle variation of in-cylinder flow. At top dead center, the measured turbulence scales of motion in the flow with the TGV closed were two thirds of those obtained with the TGV open.

The chemiluminescence emission intensity can be measured with high temporal resolution, leading to understanding the chemical reaction. Time-series chemiluminescence measurements of OH*, CH* and C2* were carried out to understand flame propagation speed, its thickness and A/F ratio of combustion status. The optical piston head (quartz) allows us to visualize combustion chamber. It is found that the chemiluminescence intensity ratio of CH*/OH* and C2*/OH* can estimate local A/F. The A/F measured by O2 sensor was used for evaluation and the results indicate this method can be applicable to estimate A/F.

We have measured the local flame front structure and its thickness using high speed three CCD camera and Developed Cassegrain optics, which could measure local OH*, CH* and C2* radicals at a point. The time-series OH*, CH* and C2* planar images are compared to those measured by local radical. The results show that the CH* and C2* signals can be a nice marker of flame front structure and its thickness, while OH* has some uncertainty.

The purpose of this study is to reveal the flame propagation characteristics. Planar OH* image and local radical emission were measured simultaneously. Planar OH* images were used to analyze the flame propagation characteristics by high-speed camera. These images were then used to evaluate the speed of distribution and the direction of flame propagation. By comparing local point radical emission and planar OH*, the flame propagation characteristics was measured and evaluate that. And the time history of the radical intensity and planar OH* distribution were compared. The relation ship between flame propagation speed and initial heat generation was discussed. The variation of flame propagation speed and the difference of propagation speed in both port sides were confirmed.

The purpose of this study was to examine the cause of fluctuations in combustion. It is important to understand the changes that occur in flame kernel development and in flame propagation during cyclic variation. In this study, a comparison was made between time-series variations in OH emission with THC concentration, and the intensity of the combustion reaction and the direction of flame propagation are also discussed. Early flame development and cyclic variation at an early stage of combustion were demonstrated by simultaneously measuring a two-dimensional image of flame emission and the time-series variation of local flame emission. The instantaneous intensity at Cassegrain measurement point agreed with the intensity of time-series variation in local flame propagation at CCD recorded timing. Variations in THC concentration in the cylinder were compared with time-series variations in local flame emission.

The purpose of this study is to examine the cause of combustion fluctuation in a partially loaded two-stroke engine with respect to the hydrocarbon (HC) concentration in the cylinder. HC concentration in the cylinder, exhaust gas velocity and pressure were simultaneously measured in order to determine the influence of HC concentration on combustion fluctuation. A correlation between cyclic variation in HC concentration in the cylinder and IMEP was confirmed. The way in which the HC concentration influenced the combustion states in the next cycle made clear. A decrease of HC concentration cause the delay of early flame development and combustion, the decrease of HC concentration had an great influence on the combustion states. The relationship between combustion states and HC concentration was discussed. The relative value of IMEP and HC concentration were closely related to the HC concentration in the cylinder.

Very high data rate Phase Doppler Measurements were carried out in order to demonstrate the spray characteristics at each cycle and how each injection differed from each other. Conventional time-averaged data analysis can hardly provide information to analysis cyclic variation of spray formation and droplet dynamics so that a cycle-resolved PDA system was developed in the study. A direct gasoline injector for two-stroke marine engine was used for the experiment. For data analysis, droplet dynamics and characteristics of different droplet diameter were examined. The results show mat cycle variation of injector was remarkable, the maximum spray tip velocity differed from 63 m/s to 93 m/s even for the consecutive injection. The data rate obtained was over 40 kHz (Max: 85 kHz) and bin width was carefully examined to show the spray collision to air and entrained air motion.

The droplet characteristics of the air-assisted gasoline injector was investigated. A Phase Doppler technique was used to measure droplet diameter and its velocity. The size-classes technique was employed and found to be the best way to understand what kind of droplet is existing in shear flow induced mushroom vortex at spray shell. The detail spray characteristics near nozzle was discussed and the double shell structure was found. The droplets of less than 20 μm can be entrained into mushroom vortex, while the larger of over 30 μm penetrates straight to downstream. The slip velocity and relative Reynolds number were used in data analysis in order to understand the momentum transfer occurrence region due to strong drag force. The spray animation was demonstrated with the ensembled / size-classified droplet, which was found to be the powerful tool to understand spray formation and dispersion process.

The purpose of this study was to detect a misfiring cycle in terms of the transfer-passage and the exhaust-pipe flow rate by experimental measurements. Simultaneous measurements of flow rates and in-cylinder pressure were carried out. The flow rate data were grouped into the different combustion classes by the in-cylinder pressure. A large flow rate of exhaust blow-down and a large reverse flow rate were observed in the cycle before misfiring, compared with in the cycle before firing. It showed that high concentration of the residual burnt gas in the cylinder was the main source of misfiring, this feature was also demonstrated by the complementary measurement of CO and CO2 concentrations.

The purpose of this study is to experimentally investigate in-cylinder flows with cyclic variation in a practical part-loaded two-stroke engine. First, the in-cylinder LDV measurements are introduced, which were carried out above the port layout and the combustion chamber as well as the exhaust pipe or the transfer port together with the simultaneous pressure measurements. Second, the in-cylinder flow characteristics in different combustion groups were discussed. The in-cylinder flow and the combustion-chamber flow were not simply characterized by the pressure variation in the engine or the other passage flow in the exhaust pipe or the transfer port. Finally, the in-cylinder flow structure with three stages was shown using the vector variation analysis and the drawing of the velocity profiles in the engine parts.

This paper reports on an experimental study of dispersion process of an air-assisted injected spray with a view of optimizing its characteristics. A phase Doppler anemometer was used to measure the injected spray characteristics under open air condition. Ensembled average mean diameters and their velocities of fuel droplets were calculated for repetitive injections. The ensembled average mean velocity showed reasonable agreement with the velocity obtained by visualization. At higher load conditions, the atomization was achieved at high speed air flow such that the mean diameter was reduced. From medium to full load conditions, the fuel droplets were distributed in a tube-like profile due the shape of the nozzle. The first group of droplets had high velocity and small mean diameter in contrast to the second group. Relative slip velocity was not so small even in the fine droplets within the air flow from the injector as can expected.

The purpose of this study is to experimentally understand the cyclic variation of combustion state in a two-stroke engine with respect to the variations in scavenging flow and the CO and CO2 emissions. The criteria of grouping combustion states into misfiring were established using the in-cylinder pressure at the crankangle of maximum variability in peak pressure instead of indicated mean effective pressure. The CO and CO2 emissions and the flow velocity variations in the transfer port and the exhaust pipe were measured. Combustion of each cycle was grouped into misfiring, incomplete firing or firing by the criteria of the in-cylinder pressure. In the cycle before misfiring, the CO and CO2 concentration showed high level and the first peak of the exhaust flow showed large velocity and the positive velocity remained for long duration, and the exhaust and the transfer port flow were steeply decelerated to negative velocity midway between scavenge port opening and bottom dead center.

Misfiring cycles were detected by a conditional sampling method to demonstrate the differences between firing and misfiring of the scavenging flow characteristics at the scavenging port and exhaust pipe using LDV method. The results show that the flow at the scavenging port was not influenced significantly by misfiring, but the blowdown flow in the exhaust pipe greatly depended on the combustion status. The blow-down flow of fired cycles at a light-load condition was very similar to the flow at a full-load condition. It was also found that measured flow characteristics at partial load should not be considered by averaging firing and misfiring cycles. The occurrence pattern of misfiring should be quantified and considered in the analysis.

The velocity variations of the burnt exhaust gas in a practical fired two-stroke engine operating under wide-open-throttle conditions were measured by a fiber LDV ( FLDV ). The characteristics of the exhaust flow are discussed in comparison with those in motoring and in a transfer port. The relation between velocity variation and pressure wave propagation in the exhaust pipe are also investigated. The measured results show that the velocity distribution in the exhaust pipe can be characterized as pulsative flow. The flow characteristics had large influence by the combustion pressure wave propagation. During exhaust and transfer-port opening, the intake flow and the blow-down flow have similar velocity gradient and peak location. The velocity distribution in the exhaust pipe was also measured, which showed pulsative flow variation having no recirculating vortex.

The flow vector variations at the transfer port exit in a small two-stroke engine under firing condition were investigated experimentally. A fiber LDV system was used to measure the two-dimensional velocities near the cylinder to obtain the scavenging flow vector. The scavenging flow vector variations at different engine speeds were discussed, and the relation between its vector behavior and the pressure differences between the exhaust pipe and the crankcase was examined. The measurement results show that the velocity profiles at the scavenging port were not uniform and to obtain the representative velocity at the port exit was impossible. But the major features of the scavenging flow can be understood from the pressure difference between the exhaust pipe and the crankcase. The start timing of the scavenging flow was delayed due to the residual gas and high pressure in the cylinder when the scavenging port was opened.

Intake flow rate of practical small two-stroke engines has been controlled by the opening ratio of throttle valve. The purpose of this study is to investigate flow behavior behind a carburetor for different throttle opening ratios in relation with the pressures. Velocities in the intake pipe were measured by a fiber LDV together with pressures at different positions under motoring and firing conditions at 3000 rpm. The results show that the recirculation vortex were formed behind the carburetor and its position and size depended on the throttle opening ratios. The variation of intake volume flow rate with the opening ratio was made clear quantitatively.

The flow characteristics of a small two-stroke engine were investigated. Direct velocity measurements were carried out by a fiber laser Doppler velocimeter (FLDV) developed under the fired conditions at engine speeds of 3000, 4000, and 5000 rpm. The first velocity peak in the firing engine was much higher than that in the motoring engine, and a negative velocity region and a large second velocity peak were observed, which were caused by the back pressure from the exhaust pipe. With increasing engine speed, the spatial velocity distribution varied in the scavenging port due to the reverse flow from the cylinder into the crankcase. The charging flow rate of the fresh air was obtained and compared with that of the motoring case.

The flow velocity in a scavenging port of a small two-stroke engine was measured directly by a specially developed fiber LDV. The measurement was carried out under motored conditions at engine speeds of 1500 to 5000 rpm, and with throttle-opening ratios of 100, 50, and 20 %. The performance of the FLDV was improved for measuring the scavenging velocity in the backscatter mode. The flow in the scavenging port changed significantly from -11 m/s to 47 m/s with the engine cycle, and the pressure difference between the crankcase and the exhaust pipe provided the typical features of the flow but not the absolute values. The results show that the scavenging flow entered the cylinder just before scavenging port opening (SO) and reached a maximum at the crank angle of 145°, which was constant for all conditions. A second velocity peak was formed by the back pressure from the exhaust pipe. The charging rate of the fresh air into the cylinder was obtained to evaluate the engine performance.