Search Results

Substantial improvements in engine fuel efficiency, torque and reduction of emissions are available with camless actuation capable of continuous control of engine valve lift, duration and timing. A phenomenological model has been developed for an unthrottled operation that is key to efficiency gain. An adaptive nonlinear controller has been designed to coordinate intake valve lift and duration by using high sampling rate intake manifold pressure and flow sensors. The driver torque demand is satisfied, while pumping losses are minimized. Simulation results for a 4 cylinder 2.0 L engine demonstrate event-to-event tracking and cylinder-to-cylinder balancing. The controller corrects for variations in effective flow areas (e.g. valve deposits), induction ram effects, and temperature.

Results of laboratory studies of the static characteristics of several different commercially available heated exhaust gas oxygen sensors are described. In these studies, the emf of the sensors was measured as a function of temperature and of the composition of calibrated gas mixtures. Several different binary gas mixtures (H2/N2, CO/N2, C3H6/N2, C3H8/N2, and CH4/N2) were used together with a variable amount of O2. In addition to laboratory studies, the same sensors were also studied in the exhaust gas of an engine. Whereas at high temperatures thermodynamic equilibrium appears to prevail, clear departures from thermodynamic equilibrium are observed at some lower temperatures (the value of which depends on the specific sensor and the specific gas mixture used). This behavior is manifested by shifts of the emf step away from stoichiometry, broadening of the step, abnormally high emf values in excess oxygen mixtures, and abnormally low emf values in reducing gas mixtures.

Quantitative measurements were made of NOx feedgas emissions during transient engine operation as the response time of an EGR system was progressively-degraded. For a simple acceleration-cruise-deceleration engine speed/torque versus time trajectory, it was found that the NOx emissions were higher during acceleration and lower during deceleration than corresponding values predicted from steady-state mapping data. The magnitude of the differences, as well as the total mass of NOx integrated over the speed/torque trajectory, all increased as the EGR response time was increased. Using a simple dynamic EGR model, NOx feedgas emissions were predicted for engine operation with a production EGR system over a 128 second portion of the FTP CVS cycle. The NOx feedgas predictions were shown to be in excellent agreement with actual emission measurements.

An engine-dynamometer study was performed to quantify the air-fuel ratio (A/F) offset between the window of a three-way catalyst (TWC) and the closed-loop control point of an exhaust gas oxygen (EGO) sensor. In this study, the effects of rpm, torque, EGR, and A/F modulation were explored along with the age of the TWC and EGO sensor. In general, it was determined that the closed-loop EGO sensor control point shifts lean as a function of increasing feedgas NOx concentration, thus causing the engine A/F to move away from the high NOx conversion efficiency regime of the TWC.

The transient response of ZrO2 and TiO2 EGO sensors has been investigated under actual engine operating conditions. The results of this study show that the response of an EGO sensor is dependent upon the characteristics of the engine and feedback control system with which it is used. Specifically, sensor response time is a function of the magnitude and frequency of the A/F changes and the initial and final values of A/F to which the sensor is exposed. ZrO2 and TiO2 sensors show similar transient behavior and have practically equivalent response times.