Flow and Combustion in a Transparent 1.9 Litre Direct Injection Diesel Engine

Author(s):  
C Arcoumanis ◽  
J H Whitelaw ◽  
W Hentschel ◽  
K-P Schindler

Two identical 1.9 litre direct injection (DI) diesel engines having optical access for application of laser diagnostics were operated at Volkswagen and Imperial College as part of the European programme (IDEA) on diesel engines. A variety of complementary laser-based techniques were used to characterize the flow-generating capacity of the intake system under steady flow conditions, the in-cylinder flow during induction and compression as well as the spray development, auto-ignition and combustion under three typical engine operating conditions. The most important results of this programme are presented and discussed here in view of their implications for improved combustion and reduction of exhaust emissions in small direct injection diesel engines, through better matching of the spray characteristics with the in-cylinder flow as a function of engine speed and load. The results were obtained in sufficient detail to allow validation of the multi-dimensional computer code developed within the IDEA programme.

2019 ◽  
Vol 9 (19) ◽  
pp. 4133 ◽  
Author(s):  
Wang ◽  
Zhang ◽  
Wang ◽  
Han ◽  
Chen

Engine knock has become the prime barrier to significantly improve power density and efficiency of the engines. To further look into the essence of the abnormal combustion, this work studies the working processes of normal combustion and knock combustion under practical engine operating conditions using a three-dimensional computation fluid dynamics (CFD) fluid software CONVERGE (Version 2.3.0, Convergent Science, Inc., Madison, USA). The results show that the tumble in the cylinder is gradually formed with the increase of the valve lift, enhances in the compression stroke and finally is broken due to the extrusion of the piston. The fuel droplets gradually evaporate and move to the intake side under the turbulent and high temperature in the cylinder. During the normal combustion process, the flame propagates faster on the intake side and it facilitates mixture in cylinder combustion. During the knock combustion simulation, the hotspots near the exhaust valve are observed, and the propagating detonation wave caused by multiple hotspots auto-ignition indicates significant effects on knock intensity of in-cylinder pressure.


Author(s):  
Lurun Zhong ◽  
Naeim A. Henein ◽  
Walter Bryzik

Advance high speed direct injection diesel engines apply high injection pressures, exhaust gas recirculation (EGR), injection timing and swirl ratios to control the combustion process in order to meet the strict emission standards. All these parameters affect, in different ways, the ignition delay (ID) which has an impact on premixed, mixing controlled and diffusion controlled combustion fractions and the resulting engine-out emissions. In this study, the authors derive a new correlation to predict the ID under the different operating conditions in advanced diesel engines. The model results are validated by experimental data in a single-cylinder, direct injection diesel engine equipped with a common rail injection system at different speeds, loads, EGR ratios and swirl ratios. Also, the model is used to predict the performance of two other diesel engines under cold starting conditions.


Author(s):  
C Arcoumanis ◽  
L N Barbaris ◽  
R I Crane ◽  
P Wisby

A cyclone-based filtration system has been developed and its potential for reduction of exhaust particulates in high-speed direct injection diesel engines is evaluated; the filtration efficiency of the four cyclones has been enhanced by means of particulate agglomeration induced by cooling in a heat exchanger. With this system installed in the exhaust pipe of a 2.5 litre direct injection engine, tests covering a wide range of speed, load and exhaust gas recirculation (EGR) fraction resulted in reductions of up to 77 per cent in emitted particulate mass flowrate. The dependence of the system's performance on engine operating conditions, EGR configuration and cyclone geometry is presented and discussed.


2017 ◽  
Vol 19 (5) ◽  
pp. 528-541 ◽  
Author(s):  
Clemens Brückner ◽  
Panagiotis Kyrtatos ◽  
Konstantinos Boulouchos

Investigations from recent years have shown that at operating conditions characterized by long ignition delays and resulting large proportions of premixed combustion, the NOx emission trend does not correspond to the (usually) postulated correlation with an appropriately defined (adiabatic) burnt flame temperature. This correlation, however, is the cornerstone of most published NOx models for direct injection diesel engines. In this light, a new phenomenological NOx model has been developed in Brückner et al. (Part 1), which considers NOx formation from products of premixed and diffusion combustion and accounts for compression heating of post-flame gases, and describes NOx formation by thermal chemistry. In this study (Part 2), the model is applied to predict NOx emissions from two medium-speed direct injection diesel engines of different size and at various operating conditions. Single parameter variations comprising sweeps of injection pressure, start of injection, load, exhaust gas recirculation rate, number of injections, and end-of-compression temperature are studied on a single-cylinder engine. In addition, different engine configurations (valve timing, turbocharger setup) and injection parameters of a marine diesel engine are investigated. For both engines and all parameter variations, the model prediction shows good agreement. Most notably, the model captures the turning point of the NOx emission trend with increasing ignition delay (first decreasing, then increasing NOx) for both engines. The differentiation in the physical treatment of the products of premixed and diffusion with increasing ignition delay showed to be essential for the model to capture the trend-reversal. Specifically, the model predicted that peak NOx formation rates in diffusion zones decrease with increasing ignition delay, whereas for the same change in ignition delay, peak formation rates in premixed zones increase. This is caused by the high energy release in short time, causing a strong compression of existing premixed combustion product zones that mix at a slower rate and have less time to mix, significantly increasing their temperature. In contrast, the model under-predicts NOx emissions for very low oxygen concentrations, in particular below 15 vol.%, which is attributed to the simple thermal NOx kinetic mechanism used. It is concluded that the new model is able to predict NOx emissions for conventional diesel combustion and for long ignition delay operating conditions, where a substantial amount of heat is released in premixed mode.


2021 ◽  
pp. 146808742110601
Author(s):  
Magnus Kircher ◽  
Emmeram Meindl ◽  
Christian Hasse

A combined experimental and numerical study is conducted on knocking combustion in turbocharged direct-injection spark-ignition engines. The experimental study is based on parameter variations in the intake-manifold temperature and pressure, as well as the air-fuel equivalence ratio. The transition between knocking and non-knocking operating conditions is studied by conducting a spark timing sweep for each operating parameter. By correlating combustion and global knock quantities, the global knock trends of the mean cycles are identified. Further insight is gained by a detailed analysis based on single cycles. The extensive experimental data is then used as an input to support numerical investigations. Based on 0D knock modeling, the global knock trends are investigated for all operation points. Taking into consideration the influence of nitric oxide on auto-ignition significantly improves the knock model prediction. Additionally, the origin of the observed cyclic variability of knock is investigated. The crank angle at knock onset in 1000 consecutive single cycles is determined using a multi-cycle 0D knock simulation based on detailed single-cycle experimental data. The overall trend is captured well by the simulation, while fluctuations are underpredicted. As one potential reason for the remaining differences of the 0D model predictions local phenomena are investigated. Therefore, 3D CFD simulations of selected operating points are performed to explore local inhomogeneities in the mixture fraction and temperature. The previously developed generalized Knock Integral Method (gKIM), which considers the detailed kinetics and turbulence-chemistry interaction of an ignition progress variable, is improved and applied. The determined influence of spark timing on the mean crank angle at knock onset agrees well with experimental data. In addition, spatially resolved information on the expected position of auto-ignition is analyzed to investigate causes of knocking combustion.


Fuel ◽  
2011 ◽  
Vol 90 (4) ◽  
pp. 1556-1566 ◽  
Author(s):  
J.V. Pastor ◽  
J.M. García-Oliver ◽  
J.M. Pastor ◽  
J.G. Ramírez-Hernández

Sign in / Sign up

Export Citation Format

Share Document