Search Results

A physics-based spark ignition model was developed and integrated into a commercial CFD code. The model predicted the spark discharge process based on the electrical parameters of the secondary ignition circuit, tracked the spark motion as it was stretched by in-cylinder gas motion, and determined the resulting energy deposition to the gas. In concert with the existing kinetic solver in the CFD code, the resulting ignition and flame propagation processes were simulated. The model results have been validated against both imaging rig experiments of the spark in moving air and against engine experimental data. The model was able to replicate the key features of the spark and to capture the cyclic variability of high-dilution combustion when multiple engine cycles were simulated.

Simulating high amounts of exhaust gas recirculation in spark ignited engines to predict combustion using the currently available CFD modeling approaches is a challenge and does not always give reasonable matches with experimental observations. One of the reasons for the mismatch lies with the secondary circuit treatment of the ignition coil and the resulting energy deposition or a complete lack of it thereof. An ignition modeling approach is developed in this work which predicts the energy transfer from the electrical circuit to the gases in the combustion chamber leading to flame kernel growth under high EGR and high gas flow velocity conditions. Secondary circuit sub-model includes secondary side of the coil, spark plug and spark gap. The sub-model calculates the delivered energy to the gas based on given circuit properties and total initial electrical energy.

Increased in-cylinder hydrogen levels have been shown to improve burn durations, combustion stability, HC emissions and knock resistance which can directly translate into enhanced engine efficiency. External fuel reformation can also be used to increase the hydrogen yield. During the High-Efficiency, Dilute Gasoline Engine (HEDGE) consortium at Southwest Research Institute (SwRI), the potential of increased hydrogen production in a dedicated-exhaust gas recirculation (D-EGR) engine was evaluated exploiting the water gas shift (WGS) and steam reformation (SR) reactions. It was found that neither approach could produce sustained hydrogen enrichment in a real exhaust environment, even while utilizing a lean-rich switching regeneration strategy. Platinum group metal (PGM) and Ni WGS catalysts were tested with a focus on hydrogen production and catalyst durability.

Many studies on low speed pre-ignition have been published to investigate the impact of fuel properties and of lubricant properties. Fuels with high aromatic content or higher distillation temperatures have been shown to increase LSPI activity. The results have also shown that oil additives such as calcium sulfonate tend to increase the occurrence of LSPI while others such as magnesium sulfonate tend to decrease the occurrence. Very few studies have varied the fuel and oil properties at the same time. This approach is useful in isolating only the impact of the oil or the fuel, but both fluids impact the LSPI behavior of the engine simultaneously. To understand how the lubricant and fuel impacts on LSPI interact, a series of LSPI tests were performed with a matrix which combined fuels and lubricants with a range of LSPI activity. This study was intended to determine if a low activity lubricant could suppress the increased LSPI from a high activity fuel, and vice versa.

Recently conducted work has been funded by the California Air Resources Board (CARB) to explore the feasibility of achieving 0.02 g/bhp-hr NOX emissions for heavy-duty on-road engines. In addition to NOX emissions, greenhouse gas (GHG), CO2 and methane emissions regulations from heavy-duty engines are also becoming more stringent. To achieve low cold-start NOX and methane emissions, the exhaust aftertreatment must be brought up to temperature quickly while keeping proper air-fuel ratio control; however, a balance between catalyst light-off and fuel penalty must be addressed to meet future CO2 emissions regulations. This paper details the work executed to improve catalyst light-off for a natural gas engine with a close-coupled and an underfloor three-way-catalyst while meeting an FTP NOX emission target of 0.02 g/bhp-hr and minimizing any fuel penalty.

To better understand real fuel effects on LSPI, a matrix was developed to vary certain chemical and physical properties of gasoline. The primary focus of the study was the impact of paraffinic, olefinic, and aromatic components upon LSPI. Secondary goals of this testing were to study the impact of ethanol content and fuel volatility as defined by the T90 temperature. The LSPI rate increased with ethanol content but was insensitive to olefin content. Additionally, increased aromatic content uniformly led to increased LSPI rates. For all blends, lower T90 temperatures resulted in decreased LSPI activity. The correlation between fuel octane (as RON or MON) suggests that octane itself does not play a role; however, the sensitivity of the fuel (RON-MON) does have some correlation with LSPI. Finally, the results of this analysis show that there is no correlation between the laminar flame speed of a fuel and the LSPI rate.

The primary focus of this investigation was to determine the hydrogen reformation, efficiency and knock mitigation benefits of methanol-fueled Dedicated EGR (D-EGR®) operation, when compared to other EGR types. A 2.0 L turbocharged port fuel injected engine was operated with internal EGR, high-pressure loop (HPL) EGR and D-EGR configurations. The internal, HPL-EGR, and D-EGR configurations were operated on neat methanol to demonstrate the relative benefit of D-EGR over other EGR types. The D-EGR configuration was also tested on high octane gasoline to highlight the differences to methanol. An additional sub-task of the work was to investigate the combustion response of these configurations. Methanol did not increase its H2 yield for a given D-EGR cylinder equivalence ratio, even though the H:C ratio of methanol is over twice typical gasoline.

It is projected that even when the entire on-road fleet of heavy-duty vehicles operating in California is compliant with 2010 emission standards of 0.20 g/bhp-hr, the National Ambient Air Quality Standards (NAAQS) requirements for ambient ozone will not be met. It is expected that further reductions in NOX emissions from the heavy-duty fleet will be required to achieve compliance with the ambient ozone requirement. To study the feasibility of further reductions, the California Air Resources Board (CARB) funded a research program to demonstrate the potential to reach 0.02 g/bhp-hr NOX emissions. This paper details the work executed to achieve this goal on the heavy-duty Federal Test Procedure (FTP) with a heavy-duty natural gas engine equipped with a three-way catalyst. A Cummins ISX-12G natural gas engine was modified and coupled with an advanced catalyst system.

The impact of additive and oil chemistry on low speed pre-ignition (LSPI) was evaluated. An additive metals matrix varied the levels of zinc dialkyldithiophosphate (ZDDP), calcium sulfonate, and molybdenum within the range of commercially available engine lubricants. A separate test matrix varied the detergent chemistry (calcium vs. magnesium), lubricant volatility, and base stock chemistry. All lubricants were evaluated on a LSPI test cycle developed by Southwest Research Institute within its Pre-Ignition Prevention Program (P3) using a GM LHU 2.0 L turbocharged GDI engine. It was observed that increasing the concentration of calcium leads to an increase in the LSPI rate. At low calcium levels, near-zero LSPI rates were observed. The addition of zinc and molybdenum additives had a negative effect on the LSPI rate; however, this was only seen at higher calcium concentrations.

In response to the sensitivity to diesel aftertreatment costs in the medium duty market, a John Deere 4045 was converted to burn gasoline with high levels of EGR. This presented some unique challenges not seen in light duty gasoline engines as the flat head and diesel adapted ports do not provide optimum in-cylinder turbulence. As the bore size increases, there is more opportunity for knock or incomplete combustion to occur. Also, the high dilution used to reduce knock slows the burn rates. In order to speed up the burn rates, various levels of swirl were investigated. A four valve head with different levels of port masking showed that increasing the swirl ratio decreased the combustion duration, but ultimately ran into high pumping work required to generate the desired swirl. A two valve head was used to overcome the breathing issue seen in the four valve head with port masking.

A series of ignition systems were evaluated for their suitability for high-EGR SI engine applications. Testing was performed in a constant-volume combustion chamber and in a single-cylinder research engine, with EGR rates of up to 40% evaluated. All of the evaluated systems were able to initiate combustion at a simulated 20% EGR level, but not all of the resulting combustion rates were adequate for stable engine operation. High energy spark discharge systems were better, and could ignite a flame at up to 40% simulated EGR, though again the combustion rates were slow relative to that required for stable engine performance. The most effective systems for stable combustion at high EGR rates were systems which created a large effective flame kernel and/or a long kernel lifetime, such as a torch-style prechamber spark plug or a corona discharge igniter.

Proposed emissions standards will require that emissions control systems function at extremely high efficiency. Recently, studies have shown that elevated gasoline fuel sulfur levels (GFSL) can impair catalytic converter efficiency. In this study, a variety of tri-metal catalysts were evaluated to determine if formulation changes could reduce emissions sensitivity to GFSL. Catalysts with elemental composition similar to an OEM, but with double the precious metal (PM) loading, were evaluated using 38 and 620 ppm GFSL. Doubling the PM loading significantly reduced catalyst sensitivity to sulfur. Doubling the rhodium loading, at the expense of the platinum loading, significantly improved NOx emission sulfur sensitivity.