Search Results

The paper presents a 1D-3D numerical model to simulate the scavenging and combustion processes in a small-size spark-ignition two-stroke engine. The engine is crankcase scavenged and can be operated with both gasoline and Natural Gas (NG). The analysis is performed with a modified version of the KIVA3V code, coupled to an in-house developed 1D model. A time-step based, two-way coupled procedure is fully described and validated against a reference test. Then, a 1D-3D simulation of the whole two-stroke engine is carried out in different operating conditions, for both gasoline and NG fuelling. Results are compared with experimental data including instantaneous pressure signals in the crankcase, in the cylinder and in the exhaust pipe. The procedure allows to characterize the scavenging process and quantify the fresh mixture short-circuiting, as well as to analyze the development of the NG combustion process for a diluted mixture, typically occurring in a two-stroke engine.

Spray impingement on walls is an important physical process in modern DI Diesel engines as it greatly influences mixture formation, combustion process and exhaust emissions. The mixture preparation is, in fact, a crucial aspect for the correct operation of the engine as it significantly affects the combustion process. In this paper three models, among the available in literature, have been selected and implemented in the KIVA-3V code. Namely, the models by O'Rourke and Amsden (OA model) [1, 2], by Bai and Gosman (BG model) [3] and by Lee et al. (LR model) [4, 5] are compared in terms of performance and capability of representing the splash phenomenon. The model capabilities are firstly tested comparing the numerical results with four sets of experimental literature data, characterized by low injection pressures. The high injection pressures of modern Diesel engines result in droplets velocities emerging from the nozzle greater than 300 m/s.

This paper discusses a partially stratified technology for engines running lean on natural gas. A single cylinder research engine has been modified to enable direct injection of a small quantity of natural gas through the spark plug to the region of the electrodes, independent of the overall lean homogeneous charge. Thus, a Partially Stratified Charge (PSC) is formed within the chamber allowing significant extension of the lean limit of combustion. Although PSC has been shown to reduce NOX emissions and improve combustion efficiency, high hydrocarbon emissions have been observed and this was thought to be due to poor mixing of the injected fuel air charge. The mixed experimental-numerical activity described herein, carried out by the Universities of British Columbia of Vancouver and Roma Tor Vergata, is aimed at improving the micro-direct injection PSC process.

In modern DI diesel engines with high pressure injection systems, the impingement of the spray on the piston head frequently occurs. Being the mixture preparation a crucial aspect for the correct operation of the engine, as it greatly influences and alters the combustion process, numerical modelling of the spray-wall interaction becomes essential. Three different spray-wall interaction models have been tested and integrated into a modified version of the KIVA-3V. All the models conserve mass, momentum and energy of the impinging droplet and have a different behavior if the surface is wet or dry. The paper focuses on the main features of the single models, investigating the different criteria used to define the splash occurrence, the ratio of the splashed mass to the incident mass and the splashed droplet size. Comparisons with experimental data are presented. The effects of the initial injection velocity and the spray cone angle are also investigated.

Ultra lean-burn natural gas engines offer the potential for lower emissions and higher efficiency than conventional SI engines. Combustion instabilities near the lean limit can be addressed by partially stratifying the in-cylinder charge. The Partially Stratified Charge (PSC) approach involves micro-direct-injection of pure fuel, or a fuel-air mixture, to create a rich zone in the region of the spark-plug. This has been demonstrated to improve combustion in an ultra-lean bulk mixture. An experimental premixing apparatus was devised to investigate the effect of changing the stoichiometry of the micro-direct-injected charge. In conjunction, a numerical methodology was used as an aid to understanding the complex in-cylinder processes. Although rich premixed micro-injection improved engine performance over the homogeneous case, the fastest heat release rate was found to occur with a pure fuel PSC charge.

The correct simulation of combustion process allows to better face several SI engines design problems, not only for innovative mixture formation concepts (stratified or ultra-lean charge), but for traditional homogeneous mixture as well. Even though many commercial codes are able to describe the complex 3-D non reacting fluid dynamics in ICE, the simulation of high turbulent flame propagation does not seem to be a completely solved problem yet. In this work a comparison between two different turbulent combustion models (a characteristic time based one by Abraham and Reitz [2, 15, 16] and a flamelet based one by Cant and AbuOrf [4, 20]) has been performed using KIVA-3V code to assess simulation reliability. Models predictive capabilities have been tested with reference to specific data acquired at the engine test bench of Tor Vergata Mechanical Engineering Department on a Fiat Punto 1242 cc 8 valves SI engine over a wide range of operating conditions.

The design of a “high-performance” airbox for a naturally aspirated internal combustion engine (ICE) of a car racing in prototype sport competitions is described. A computational approach to achieve optimum airbox geometry in terms of fluid dynamical losses reduction and engine volumetric efficiency improvement is proposed. Experiments on race track have been carried out to test the car performances improvement. The numerical calculations have been done using a 3D numerical code. The code solves finite-difference approximation of the fluid dynamic governing equations (continuity, momentum and energy balance). The solution has been performed numerically by an integration both in space and time by means of the Arbitrarian Lagrangian Eulerian (ALE) technique. The numerical simulations have been carried out imposing steady and unsteady boundary conditions.

The performances increasing of internal combustion engines for race car has driven to develop special systems in order to improve the volumetric efficiency. To this aim, in the last years, a great effort has been done especially in studying geometries for airbox, turbo-compressors, special exhaust systems, etc. In this paper, the project of a “high-performance” airbox for a naturally aspirated internal combustion engine (ICE) of a car racing in prototype sport competitions is described. In order to optimize the airbox geometry under extremely complex operative conditions, the fluid dynamic phenomena inside the airbox have been studied by means of a three dimensional computational code (3D CFD). This approach has allowed to study different airbox geometry and to define the one to be realized and tested on race track. The new airbox geometry, defined in this way, has brought to good results.

The use of Compressed Natural Gas as an alternative fuel in urban transportation is nearly established and represents an efficient short and medium term solution to face with urban air pollution. However, in order to completely exploit its potential, the engine needs to be specifically designed to operate with this fuel. In the latest years, the authors have investigated the performances of a Heavy Duty Turbocharged CNG fuelled engine both experimentally and by using some analytical tools specifically developed by them which have been used for the engine optimisation. In the present paper the simulation approach has been enlarged by means of a co-operative use of a CFD code and experimental analysis on the actual engine. The numerical simulation of combustion process has, in fact, been used, to interpret series of pressure cycles, aiming to analyse how cyclic fluctuations influence engine behaviour in terms of combustion efficiency and temperature and pollutant distribution.

The first part of the paper gives an overview of the environmental conditions with which a future two stroke powered vehicle must comply and explains the reasons for which a direct gasoline injection into the combustion chamber offers a potential solution. The paper continues with a description of the fuel/air mixture injection used in the F.A.S.T. concept and gives a detailed overview of the layout of the 125 cc engine to which it is applied. The structure of its electronic engine management system, mandatory for the necessary control precision, is presented. Hereafter is made a short introduction to the visualization and numerical computation tools used for the engine design optimization. The paper concludes with a detailed presentation and discussion of the experimental results obtained with the engine operated, either in steady state and transient conditions on an engine test rig, and mounted in a classic small dimension two-wheel vehicle submitted to road tests.