Search Results

A quasidimensional engine simulation which uses the concepts of fractal geometry to model the effects of turbulence on flame propagation in a homogeneous charge SI engine has been developed. Heat transfer and blowby/crevice flow submodels are included in this code and the submodels chosen are found to be reasonable. The model predictions of cylinder pressure histories are then compared with experimental data over a range of loads, equivalence ratios, and engine speeds. The model is not adjusted in any manner to yield better agreement with the data, other than by tuning the simple turbulence model used so as to yield agreement with data for the nonreacting flow. However, current information about the flame wrinkling scales in an engine is inadequate. Therefore, predictions are made for three different assumptions about the flame wrinkling scales which span the range of physically possible scales.

An investigation of cyclic variability in a spark ignition engine is reported. Specifically, the predictions of an engine code have been compared with experimental data obtained using a well-characterized SI engine. The engine used for the experimental work and modeled in the code is the single cylinder research engine developed at Sandia National Laboratories and now operating at Drexel University. The data used for comparison were cylinder pressure histories for 110 engine cycles gathered during operation at a single engine operating condition. The code allows the various factors that could influence cyclic variability to be examined independently. Specifically, a model has been used to independently examine the effects of variations in equivalence ratio and of the turbulence intensity on cycle-to-cycle variations in the peak cylinder pressure, the crankangle of occurrence of peak pressure, the flame development angle, and the rapid burning angle.

Condensation of fuel vapor on the cold surfaces within the combustion chamber is investigated as a possible mechanism for increased HC emissions from SI engines during cold start. A one-dimensional, transient, mass diffusion analysis is used to examine the condensation of single-species fuels on the surfaces of the combustion chamber as the pressure within the cylinder rises during compression and combustion, and re-vaporization during expansion, blowdown, and exhaust. The effects of wall temperature, fuel volatility, and engine load and speed on this mechanism are also discussed. This analysis shows that low-volatility fuel components can condense on the surfaces of the combustion chamber when the surface temperatures are sufficiently low. This condensed fuel may re-vaporize during the power and exhaust strokes, or it may remain in the combustion chamber until surface temperatures rise, perhaps tens of seconds later.

This paper presents the results of an experimental investigation of the fuel-air mixing process in a port-fuel-injected, 4-valve, spark-ignited engine that was motored to simulate cold cranking and start-up conditions. An infrared fiber-optic instrumented spark plug probe was used to measure the local, crank angle resolved, fuel concentration in the vicinity of the spark gap of a single-cylinder research engine with a production head and fuel injector. The crank-angle resolved fuel concentrations were compared for various injection timings including open-intake-valve (OIV) and closed-intake-valve (CIV) injection, using federal certification gasoline. In addition, the effects of speed, intake manifold pressure, and injected fuel mass were examined.

The basic process of spark ignition in engines has changed little over the more than 100 years since its first application. The rapid evolution of several advanced engine concepts and the refinement of existing engine designs, especially applications of power boost technology, have led to a renewed interest in advanced spark ignition concepts. The increasingly large rates of in-cylinder dilution via EGR and ultra-lean operation, combined with increases in boost pressures are placing new demands on spark ignition systems. The challenge is to achieve strong and consistent ignition of the in-cylinder mixture in every cycle, to meet performance and emissions goals while maintaining or improving the durability of ignitor. The application of railplug ignition to some of these engine systems is seen as a potential alternative to conventional spark ignition systems that may lead to improved ignition performance.

A railplug is a new type of ignitor for SI engines. A model for optimizing energy deposition in a railplug ignition system is developed. The model is experimentally validated using a low voltage railplug ignition circuit. The effect of various ignition circuit parameters on the energy deposition and its rate are discussed. Durability of railplugs is an important factor in railplug circuit design. As for all spark ignitors, durability of a railplug decreases as energy deposition is increased. Therefore recommendations are made to minimize wear and increase durability, while depositing sufficient energy to attain ignition, using a railplug.

A fiber optic instrumented spark plug was used to make time-resolved measurements of the fuel vapor concentration history near the spark gap in a four-valve DISI engine. Four different bulk flow were investigated. Several early and late injection timings were examined. The fuel concentration at the spark gap was correlated with IMEP. Emissions of CO, HCs, and NOx were related to the type of bulk flow. For both early and late injection the CoVs of fuel concentration were generally lowest for the weakest bulk flow which resulted in a stable stratification. Strong bulk flows convected the inhomogeneities through the measurement area near the spark plug resulting in both large intracycle and cycle-to-cycle variation in equivalence ratio at the time of ignition.

Improved submodels for use in a dynamic engine/vehicle model have been developed and the resulting code has been used to analyze the tip-in, tip-out behavior of a computer-controlled port fuel injected SI engine. This code consists of four submodels. The intake simulation submodel is similar to prior intake models, but some refinements have been made to the fuel flow model to more properly simulate a timed port injection system, and it is believed that these refinements may be of general interest. A general purpose engine simulation code has been used as a subroutine for the cycle simulation submodel. A conventional vehicle simulation submodel is also included in the model formulation. Perhaps most importantly, a submodel has been developed that explicitly simulates the response of the on-board computer (ECM) control system.

The mixture preparation process was investigated in a direct-injected, 4-valve, SI engine under motored conditions. The interaction between the high-pressure fuel jet and the intake air-flow was observed. Laser-sheet droplet imaging was used to visualize the in-cylinder droplet distributions, and a single-component LDV system was used to measure in-cylinder velocities. The fuel spray was visualized with the engine motored at 1500 and 750 rpm, and with the engine stopped. It was observed that the shape of the fuel spray was distorted by the in-cylinder air motion generated by the intake air flow, and that this effect became more pronounced with increasing engine speed. Velocity measurements were made at five locations on the symmetry plane of the cylinder, with the engine motored at 750 rpm. Comparison of these measurements with, and without, injection revealed that the in-cylinder charge motion was significantly altered by the injection event.

Predictions of the Fractal Engine Simulation code were compared with SI engine data in a previous paper. These comparisons were extremely good except for the single data set available at a low engine speed. Because of uncertainty regarding whether the lack of agreement for this case resulted from some difficulty with the experimental data or was due to lack of proper speed dependence in the model, additional comparisons are made for a range of speeds from 300-1500 rpm. The fractal burning model is a turbulence driven model (i.e., driven primarily by the turbulence intensity) that divides the combustion process into four sequential phases: 1) kernel formation, 2) early flame growth, 3) fully developed turbulent flame propagation, and 4) end of combustion. The kernel formation process was not included in the previous version of this model, but was found to be required to predict engine speed effects.

Multi-dimensional numerical simulation of the combustion process in spark ignition engines were performed using the Coherent Flame Model (CFM) which is based on the flamelet assumption. The CFM uses a balance equation for the flame surface area to simulate flame surface advection, diffusion, production and destruction in a turbulent reacting flow. There are two model constants in CFM, one associated with the modeling of flame surface production and the other with the modeling of flame surface destruction. Previous experimental results on two test engines charged with propane-air mixtures were used to compare with the computations for different engine speeds, loads, equivalence ratios and spark plug locations. Predicted engine cylinder pressure histories agree well with the experimental results for various operating conditions after the model constants were calibrated against a reference operating condition.

The results of continuing investigations of a new type of ignitor, the railplug, are reported. Previous studies have shown that railplugs can produce a high velocity jet that is driven both by electromagnetic and thermal forces and that the jet velocity is strongly affected by the railplug geometry and by the electronics characteristics of the follow-on circuit. The present research was intended to provide insights about both: 1) how to match the electronics characteristics to a given geometry and 2) how the geometry affects the jet velocity. It is found that faster current rise times result in higher plasma velocities but current pulses that are too short result in rapid deceleration of the plasma while it is still within the railplug. It is also found that a fundamental geometric parameter is the ratio of the inductance gradient to the volume trapped within the railplug: the larger L′/V, the faster the resulting combustion process.

Researchers in the field of turbulent combustion have found fractal geometry to be a useful tool for describing and quantifying the nature of turbulent flames. This paper describes and compares several techniques for the fractal analysis of two dimensional (2-D) turbulent flame images. Four methods of fractal analysis were evaluated: the Area Method, the Box Method, the Caliper Method, and the Area-Caliper Method. These techniques were first applied to a computer-generated fractal image having a known fractal dimension and known cut-offs. It was found that a “window” effect can cause the outer cut-off to be underestimated. The Caliper Method was found to suffer from noise arising from the statistical nature of the analysis. The Area-Caliper Method was found to be superior to the other methods. The techniques were applied to two types of flame images obtained in a spark ignition engine: Mie scattering from particles seeded in the flow and laser induced fluorescence of OH.

A new type of ignitor, the railplug, shows promise of extending the dilution limits for spark ignition engines. While much of the effort expended in our study of railplugs has focused upon demonstrating their effectiveness, it is recognized that railplug durability is presently not acceptable for production engine applications. The goal of the present study was to examine the factors that affect durability. The results of two types of investigations are reported. The effects of rail materials, pressure, delivered energy, and voltage at constant delivered energy on electrode erosion rates were studied for repeated firings in air at constant pressure. Railplug durability in a four-stroke SI engine was also evaluated, including examination of the effects of delivered energy, current pulse characteristics, and materials.

We present a multidimensional numerical model that calculates turbulent premixed flame propagation, assuming the flames have fractal geometries. Two scaling transformations, previously developed for laminar flames, are used to incorporate the fractal burning model in KIVA-II1, a numerical hydrodynamics code for chemically reactive flows. In this work the model is implemented for propane/air mixtures. For applications to internal combustion engines, we have also developed a fractal model for early flame kernel growth. Our multidimensional model can be used in experimental comparisons to test postulated fractal parameters, and we begin this task by comparing calculated results with measurements of propane/air combustion in a spark ignition engine. Good agreement is obtained between computed and measured flame positions and pressures in all cases except a low engine speed case.

Lean mixture combustion might be an important feature in the next generation of SI engines, while diluents (internal and external EGR) have already played a key role in the reductions of emissions and fuel consumption. Lean burn modeling is even more important for engine modeling tools which are sometimes used for new engine development. The effect of flame strain on flame speed is believed to be significant, especially under lean mixture conditions. Current quasi-dimensional engine models usually do not include flame strain effects and tend to predict burn rate which is too high under lean burn conditions. An attempt was made to model flame strain effects in quasi-dimensional SI engine models. The Ford model GESIM (stands for General Engine SIMulation) was used as the platform. A new strain rate model was developed with the Lewis number effect included.

In premixed turbulent combustion models, two mechanisms have been used to explain the increase in the flame speed due to the turbulence. The newer explanation considers the full range of turbulence scales which wrinkle the flame front so as to increase the flame front area and, thus, the flame propagation speed. The fractal combustion model is an example of this concept. The older mechanism assumes that turbulence enables the penetration of unburned mixtures across the flame front via entrainment into the burned mixture zone. The entrainment combustion or eddy burning model is an example of this mechanism. The results of experimental studies of combustion regimes and the flame structures in SI engines has confirmed that most combustion takes place at the wrinkled flame front with additional combustion taking place in the form of flame fingers or peninsulas.

The Texas Project involves the conversion of light-duty vehicles, up to and heavy light-duty trucks, to bi-fueled vehicles using commercially available aftermarket CNG and LPG conversion systems. The test fleet includes 68 dual fueled conversions. Virtually every type of aftermarket conversion technology for CNG and LPG was evaluated: eight different CNG and seven different LPG conversion “kits”, all of which are modern systems incorporating closed-loop control. The kits were installed and calibrated according to the manufacturer's guidelines and recommendations. The emissions when operating on the alternative fuel were compared to those when operating on certification gasoline to determine the “success” of the conversion. Many of these conversions, performed according to the manufacturer's requirements, were not “successful” (worse emissions than for gasoline operation). In almost all cases, the problem was NOx emissions that were too high when operating on the alternative fuel.

A reduced autoignition mechanism for iso-octane has been developed by identifying paths to formation of the stable species measured during motoring knock experiments and eliminating paths to formation of species that were not measured. The resulting mechanism includes low-, intermediate-, and high-temperature reactions and consists of 103 species and 131 reactions. This mechanism differs from detailed models not only in the number of reactions and species, but most importantly, in the nature and rates of the degenerate chain branching reactions. To implement this mechanism, a knock subroutine has been added to a quasidimensional spark ignition engine model that accounts for heat losses, blowby, etc. Thus, errors in the reactivity predictions can be assigned almost exclusively to the kinetics. Numerical predictions of the exhaust composition during motoring knock are compared with experimental measurements as a function of compression ratio for several operating conditions.

Fuel transport was visualized within the cylinder of a port injected four-valve SI engine having a transparent cylinder liner. Measurements were made while motoring at 250 rpm to simulate cranking conditions prior to the first firing cycle, and at 750 rpm to examine the effects of engine speed. A production GM Quad-4 cylinder head was used, and the stock single-jet port fuel injector was used to inject indolene. A digital camera was used to capture back-lighted images of cylinder wall wetting for open and closed intake valve injection. In addition, two-dimensional planar imaging of Mie scattering from the indolene fuel droplets was used to characterize the fuel droplet distribution as a function of crank angle for open and closed intake valve injection. LDV was used to measure the droplet and air velocities near the intake valves during fuel induction. It was found that with open-valve injection a large fraction of the fuel impinged on the cylinder wall opposite the intake valves.