Search Results

Intake flow simulations were carried out for a prototype DISI engine using the standard k-ε model and the RNG k-ε model. The results were compared with PTV (transient water analog) measurements. The study was focused on low load operations with engine speed at 400 rev/min. Two cases were studied, a single intake case in which one intake port was blocked and a dual intake port case. In the computations, the results show that the standard k-ε model tends to produce higher turbulence levels when turbulence is generated and decays faster when turbulence dissipates. Different turbulence models predict almost the same flow structures. However, the effects of the turbulence model on the predicted tumble and swirl ratios are significant. The TKE distributions at BDC predicted by the two models are also different. The standard k-ε model seems to be more diffusive. Good agreements with PTV data were obtained in the single valve case with the RNG k-ε model.

Intense competition and global regulations in the automotive industry has placed unprecedented demands on the performance, efficiency, and emissions of today's IC engines. The success or failure of a new engine design to meet these often-conflicting requirements is primarily dictated by its capability to provide minimal restriction for the inducted and exhausted flow and by its capability to generate strong large-scale in-cylinder motion. The first criterion is directly linked to power performance of the engine, while the latter has been shown to control the burn rate in IC engines. Enhanced burn rates are favorable to engine efficiency and partial load performance. CFD based Numerical Simulations have recently made it possible to study the development of such engine flows in great details. However, they offer little guidance for modifying the ports and chamber geometry controlling the flow to meet the desired performance.

Accurate prediction of engine combustion characteristics, especially burn rates, can eliminate a number of hardware iterations, thus resulting in a significant reduction in design and developmental time and cost. An analytical methodology has been developed which allows the determination of part-load MBT spark timing to within 2 crank-angle degrees. The design methodology employs the in-house-developed steady-state quasi-dimensional engine simulation model (GESIM), coupled with full-field measurement of the in-cylinder fluid motion at bottom dead center (BDC) in the computer-controlled water analog system (AquaDyne). The in-cylinder flow-field measurements are obtained using 3-D Particle Tracking Velocimetry (3-D PTV), also developed in-house. In this methodology, the in-cylinder flow measurement data are used to calibrate both the tumble and swirl models in GESIM.

Measurements of the instantaneous in-cylinder flow fields were carried out in a water analog engine simulation rig using 3-D Particle Tracking Velocimetry[1]. Two different configurations based on a 4-valve per cylinder engine with a typical pent-roof type combustion chamber were investigated. Measurements were performed at the end of intake stroke at BDC for simulated idle conditions. Efforts were made to maximize the particle seeding density (and the resulting number of 3-D velocity vectors) to yield 250 to 300 instantaneous vectors at each cycle. Using an appropriately designed spatial filter the low pass filtered (large scale) instantaneous velocity fields were reconstructed for each cycle. The spatial cut-off of the filter was set to 1/3 of the engine bore diameter. Based on these cycle-resolved flow fields, the cycle-to-cycle variations as well as the cycle resolved turbulence fluctuations were computed.

A new methodology for rapidly characterizing the in-cylinder flow field at spark ignition for internal combustion engines is described in this paper. The process involves the use of 3-D particle tracking velocimetry to measure the flow field at intake valve closing (IVC) in a water analog engine simulation, and the use of CFD to compute the evolution of the measured flow field during the compression stroke, by using the experimental 3-D PTV results at IVC as the initial condition for the calculations. The technique has been applied to investigate the in-cylinder flow field of a typical 4 valve engine operating in two different modes; one or two intake ports active. The results indicate that in either mode the flow field at IVC is dominated by a different large scale structure: tumble in the case where both intake ports are active and swirl in the case where only one port is active. The results also indicate that these structures evolve differently during the compression stroke.

The feasibility of using experimentally determined flow fields at intake valve closing, IVC, as initial conditions for computing the in-cylinder flow dynamics during the compression stroke is demonstrated by means of a computer simulation of the overall approach. A commercial CFD code, STAR-CD, was used for this purpose. The study involved two steps. First, in order to establish a basis for comparison, the in-cylinder flow field throughout the intake and compression strokes, from intake valve opening, IVO, to top dead center, TDC, was computed for a simple engine geometry. Second, experimental initial conditions were simulated by randomly selecting and perturbing a set of velocity vectors from the computed flow field at IVC.

Flow fields generated during the intake stroke of a 4-stroke I.C. engine are studied experimentally using water analog simulation. The fluid is seeded by small flow tracer particles and imaged by two digital cameras at BDC. Using a 3-D Particle Tracking Velocimetry technique recently developed, the 3-D motion of these flow tracers is determined in a completely automated way using sophisticated image processing and PTV algorithms. The resulting 3-D velocity fields are ensemble averaged over a large number of successive cycles to determine the mean characteristics of the flow field as well as to estimate the turbulent fluctuations. This novel technique was applied to three different cylinder head configurations. Each configuration was run for conditions simulating idle operation two different ways: first with both inlet ports open and second with only the primary port open.