Analyzing different combustion chambers in gasoline engine with three dimensional modeling

An internal combustion engine (ICE) is a heat engine where the combustion of a fuel occurs with an oxidizer (usually air) in a combustion chamber that is an integral part of the working fluid flow circuit. In an internal combustion engine the expansion of the high-temperature and high-pressure gases produced by combustion apply direct force to some component of the engine. ICEs are usually powered by energy-dense fuels such as gasoline or diesel, liquids derived from fossil fuels. While there are many stationary applications, most ICEs are used in mobile applications and are the dominant power supply for vehicles such as cars, aircraft, and boats. The studies , designs and the experimental work have been carried out to improve the internal combustion engines and bring out new technologies are highly time and cost consuming, laborious work to accomplish. To have a previous step for these laborious work and predict the potential problems can occur, real-like physical and mathematical models are being constituted and these models are getting improved day by day. These models which are using finite volume method in internal combustion engines are based on computational fluid dynamics (CFD), combustion physics and chemistry. Combustion chamber divided into finite volumes and the conversation equations of mass, momentum and energy are applied to these smaller volumes. Especially in recent years, three-dimensional models of internal combustion engines began to be used frequently. With the help of these models, improving the flow in the cylinder, air movements and air-fuel mixture formation, fuel spray dynamics and fuel injector design, improvement of combustion, reducing fuel consumptions and down-sizing, supercharging and turbocharging systems, providing recent emission standarts, heat transfers and durability of engine parts, design of power transmission systems etc. studies and improvements can be performed. Although there are a number of commercial CFD packages available on the market, the most widely used for engine research has been the KIVA family of three-dimensional programs. This popularity is based on the accuracy of KIVA, and the fact that the complete source code is available to a worldwide user community at a modest cost. In addition to their use in industry and government laboratories, KIVA programs are widely used in university engineering departments, which are graduating a new generation of engineers familiar with KIVA and CFD modeling in general. KIVA3V retains the distinct three-part structure of KIVA3, in which the grid generator and graphics are separated from the hydro program. The KIVA3V package includes both a pre-processor (K3PREP), which has been expanded to support the generation of grids with valves, but in this study ICEM CFD software is used because with this software more detailed grid structures can be carried out visually. For a post-processor (K3POST) features KIVA3V software, similarly TECPLOT post-processor software was used instead of K3POST due to visuality and allows the detailed study. Air motion, air-fuel mixture formation and gas phase chemistry inside the cylinder of an IC-engine are the basic processes that describe physical and chemical steps of the power generation respectively. Both experimental and numerical studies are done by various number of engine researchers to develop more efficient engines which contribute less to air pollution. To obtain reliable numerical results, the phenomena of turbulence, mixture formation and chemical kinetics must be modeled within reasonable accuracy in a combustion chamber under transient conditions. These processes have been investigated by many researchers to obtain a mathematical model. This study is performed on three different combustion chambers. First combustion chamber has plain head piston (disk), the second combustion chamber has cylindrical bowled piston and the third combustion chamber is W type bowled and assymmetrical slotted chamber. Firstly in this study, grid structures are built as suitable for KIVA3V can solve. To build grid structures ICEM CFD software is used. Four different models are created as one of them is a second type combustion chamber with 9,1 compression ratio and three others are each different combustion chambers with 10,5 compression ratio. Experimentally only the data of second type combustion chamber with 9,1 compression ratio are present. The experiments of geometries with 10,5 compression ratio will be accomplished in later studies. The experiments performed with 9,1 compression ratio had made for a previous doctorate study in Automative Laboratories of Mechanical Engineering Faculty, ITU. Primarily correspondence of experimental and simulation results from the combustion chamber with 9,1 compression ratio is aimed. To obtain this correspondence, KIVA3V parameteres and the cell numbers at grid structure are arranged and the calibration of the code is done. Basic parameters are determined after observation of the results from calibration work and the literature review; then the phase of determining the cell numbers and combustion rate calibration is started. It is obtained that the simulation results differ a lot with respect to changing cell numbers roughly to precisely. The literature review, the computer power and the resolution process are taken into consideration and it is determined with around 200000 cell numbers the real simulations are carried out. After calibration simulations it is obtained that the combustion rates and spark advance are also need to be adjusted. The frequency factor in Arrhenius reaction rate equation that determines the combustion rate, has to be arranged for each grid structure and amount of mesh size. The only way to arrange is to make consecutive simulations and obtaining the best value. Moreover it is obtained that the ignitation delay durations of experiments and simulations show difference, therefore in simulations updated values are used instead of using the ignitation delay from experiments. In the simulations using 10,5 compression ratio, the direct values determined from calibration results are used. With standard KIVA3V code, in-cylinder pressure with respect to crank angle, average and maximum temperatures, turbulent kinetic energies, fuel amount difference, released heat, arised CO2, H2O, CO and NO amounts are examined. For three-dimensional graphics, Tecplot software is used. The solution results of the KIVA3V software are high degree of dependence on the number of cells and grid structure has been confirmed. Solution results become dramatically different with changing number of cells coarse to fine. Therefore, for a model to be established of the dimensions like in this study, it was understood that it should be used minimum 50000 number of cells and lower 2 mm cell sizes by considering the studies in the literature and as well as experienced, and also considering processing speed of computers. According to single-step oxidation reaction supplied in the KIVA3V standard software, for combustion of the gasoline has been seen that develop very suddenly. Reason of this Arrhenius reaction rate equation frequency factor "a" value that causes the fuel to burn very quickly. Therefore, for each simulation, this value had to be adjusted by experimentally. According to this result, the maximum pressure and the time they are incurred of the three combustion chambers, for 1st combustion chamber reaction rate was increased and for 3rd combustion chamber reaction rate was decreased. The air movements and the turbulent kinetic energies are focused at the bowl volume in bowled combustion chambers. On the other hand in chambers with plain head piston the focus is more at the cylinder wall. Therefore after top dead center, the heat release occurs sooner in bowled geometries, where the heat release occurs later at the plain head pistons. However the Arrhenius frequency coefficient is higher at the plain headed pistons , hence at the main combustion phase the heat release rate is higher and it causes the pressure to increase faster. At the third piston the aimed swirl movement with using the slotted piston couldn't be obtained. Accordingly the results from the third chamber showed similarities with the second chamber.