Investigation of Causes and Preventive Measures for Destructive Abnormal Combustion (Superknock) in
High-Performance Spark
Ignited Gasoline Engines, 03-R8049
Principal Investigators
Manfred Amann
Darius Mehta
Terry Alger
Inclusive Dates: 04/01/09 – 03/31/10
Background - The spark ignition (SI) engine has been known to exhibit several different abnormal combustion phenomena, most which can be addressed with improved engine design or control schemes. However, in the high performance SI engines — where turbocharging is employed to increase brake power at low engine speeds for improved driveability — a new phenomenon of abnormal combustion, described as Low-Speed Preignition (LSPI), has been exhibited. LSPI is characterized as preignition typically followed by heavy knock, which can result in severe engine damage. However, because LSPI events occur only sporadically and in an uncontrolled fashion, the causes for this phenomenon have not been successfully explained and solutions to suppress it have not been developed.
Approach - SwRI has been developing gasoline engine technologies that both improve efficiency and reduce emissions. Based on this expertise, this project investigated the causes of and potential preventive measures for LSPI.
Accomplishments - As a possible measure to suppress LSPI, the potential benefits of operating the engine with various levels of exhaust gas recirculation (EGR) have been explored. Engine tests indicate that light to moderate levels of EGR (5 to 10 percent) reduce the incidence of LSPI dramatically. By operating the engine with 10 percent EGR, an increase in fueling rates of 15 percent and engine load/ BMEP of 17 percent was possible while reducing LSPI frequency by roughly 80 percent and LSPI intensity by approximately 30 percent over non-EGR conditions. Second, several engine controls factors and operating conditions were investigated with respect to their effects on LSPI. It was recognized that the two dominant factors in influencing the LSPI occurrence frequency are engine load as governed by fueling rate (energy flux) and in-cylinder air-to-fuel ratio. When maintaining a constant fueling rate (not BMEP or torque), all other factors such as spark timing, MAT, coolant temp, etc., only played a minor, but not necessarily eliding, role in their effect on LSPI activity. From exhaust emission measurements it was also recognized that a spike in HC emissions and a significant increase in lambda was associated with LSPI. Furthermore, when inducing a LSPI-like combustion event by using large spark advance for a short duration, HC emission and exhaust port lambda were significantly lower than during "true" LSPI events. It was concluded that a hydrocarbon-based accumulation occurs in the combustion chamber over time. These additional HC are consumed during LSPI events. When combined with the results of other researchers in this field one might conclude that the leading cause of LSPI is lubricant and/or fuel-based HC accumulation in the top land piston crevices volume. To investigate the source for LSPI from a fuels perspective, four gasoline fuel blends with similar properties, such as octane rating, boiling point distribution and RVP, but vastly different compositions, were tested for their effects on LSPI in a modern turbo-charged, DI gasoline engine. From this study, it was recognized that the fuel chemical composition strongly influences the likelihood and severity of LSPI. Fuel blends with high levels of aromatics increase the frequency at which LSPI occurs somewhat where as oxygenated fuels and, especially, low aromatic blends reduced the LSPI frequency. It was also learned that despite very similar RON and MON ratings, the knock and auto-ignition characteristics of the test fuels in the DISI engine were different. In particular the low-aromatics fuel blends showed an increase tendency to auto-ignition and knock characterized by the presence of a low-temperature heat release regime prior to the main combustion phase.
