Search Results

The effects of port fuel injection (PFI) timing and targeting on air/fuel (A/F) control, exhaust emissions, and combustion stability at retarded spark timing were investigated on a 2.0L I-4 engine with production injectors (300-350 micron SMD droplet spray). Timings were fully closed valve injection (CVI) or fully open valve injection (OVI), and selected targetings were towards the valve or port floor. An “ideal” pre-vaporized, pre-mixed fuel system was also tested to provide a baseline with which to isolate PFI fuel preparation effects. The key findings were: Transient A/F excursions with PFI were minimized over the full temperature range with OVI timing and valve targeting. The X-tau modeled film mass for OVI/valve target was 50% less than CVI/valve target and 30% less than OVI/port target with a cold engine (20° C). When fully warm (90° C), the A/F response of CVI/valve target improved to near that of OVI.

Spark retard is desirable for decreasing cold start hydrocarbon emissions and lighting off the catalyst more rapidly. The focus of this work is to better understand the nature of the HC emissions as spark is retarded and investigate the location of the oxidation (in-cylinder or in the exhaust port and manifold). Fast FID measurements were taken in the exhaust port of a single cylinder research engine during cold, retarded spark engine operation (1200 rpm, 2.5 bar IMEP, 20 °C fluids). At moderate spark retard both Fast FID (exhaust port) and exhaust plenum HC levels decreased due to reduced crevice volume fraction at the end of burn, and increased in-cylinder burn up. In contrast, at large spark retard the port HC's increased dramatically while the exhaust plenum levels continued to fall to near zero. This is thought to be due to the onset of incomplete in-cylinder combustion along with increased exhaust port and manifold after-burning caused by the increasing exhaust gas temperatures.

Engine operation with spark ignition retarded from MBT timing is used at cold start to reduce HC emissions and increase exhaust gas temperature; however it also results in increased cyclic variations. Steady-state cold fluids testing was performed to better understand the causes of the cycle-to-cycle variations. Detailed analysis of individual cycles was performed to help gain an understanding of the causes of cyclic variations. The important results were: The primary cause of cyclic variations in IMEP is variations in the combustion phasing (location of 50% mass fraction burned). The expansion ratio decreases rapidly during combustion for retarded spark timing and therefore the phasing determines individual cycle thermal efficiency and IMEP. Variations in the late burn have little impact on the IMEP as this combustion occurs close to EVO and does little expansion work.

Using a fast-sampling valve, residual-fraction levels were determined in a 2.0L spark-ignited production engine, over varying engine operating conditions. Individual samples for each operating condition were analyzed by gas-chromatography which allowed for the determination of in-cylinder CO and CO2 levels. Through a comparison of in-cylinder measurement and exhaust data measurements, residual molar fraction (RMF) levels were determined and compared to analytical results. Analytical calculations were performed using the General Engine SIMulation (GESIM) which is a steady state quasi-dimensional engine combustion cycle simulation. Analytical RMF levels, for identical engine operating conditions, were compared to the experimental results as well as a sensitivity study on wave-dynamics and heat transfer on the analytically predicted RMF. Similarly, theoretical and experimental NOx emissions were compared and production sensitivity on RMF levels explored.

A new reaction scheme for NOx production is incorporated into a steady state quasi-dimensional engine combustion simulation. The reaction kinetics includes 67 reactions and 13 chemical species, and assumes equilibrium concentration for all other chemical species. The General Engine SIMulation (GESIM) developed by Ford Motor Company is used to model the engine cycle. The new reaction scheme is a super-extended Zel'dovich mechanism (SEZM) which predicts NOx formation levels to within 10% of engine test data for several engines, whereas the 3 reaction, extended Zel'dovich mechanism (EZM) is shown to have errors of approximately 50% or more for similar conditions. Analytical engine mapping, under NOx constrained calibration, requires accurate modeling of NOx emissions over varying engine operating conditions.