Pilot injection (PI) during the negative-valve-overl
ap (NVO) period is one method to improve control of combustion in gasoline controlled auto-ignition engines. This is generally attributed to both chemical and thermal effects. However, there are little experimental data on active species formed by the combusting PI and their effect on main combustion in real engines. Thus, it is the objective of the current study to apply and assess optical in-cylinder diagnostics for these species. Firstly, the occurrence and nature of combustion during the NVO period is investigated by spectrally-resolved multi-species flame luminescence measurements. OH*, CH*, HCO*, CO-continuum chemiluminescence, and soot luminosity are recorded. Secondly, spectrally-, spatially-, and cycle-resolved laser-induced fluorescence measurements of formaldehyde are conducted. It is attempted to find a cycle-resolved measure of the chemical effect of PI. However, not any noticeable correlation between the measured formaldehyde content of the charge and main combustion phasing is observed. This can be explained by the findings that the exhaust gas of the prior cycle and potentially pool fire contribute significantly to the measured formaldehyde content of the charge. Thus, formaldehyde is not a good indicator of the chemical effect of PI on main combustion under these conditions.
Pilot injection (PI) during the negative-valve-overlap (NVO) period is one method to improve control of combustion in gasoline controlled auto-ignition engines. This is generally attributed to both chemical and thermal effects. However, there are little experimental data on active species formed by the combusting PI and their effect on main combustion in real engines. Thus, it is the objective of the current study to apply and assess optical in-cylinder diagnostics for these species. Firstly, the occurrence and nature of combustion during the NVO period is investigated by spectrally-resolved multi-species flame luminescence measurements. OH*, CH*, HCO*, CO-continuum chemiluminescence, and soot luminosity are recorded. Secondly, spectrally-, spatially-, and cycle-resolved laser-induced fluorescence measurements of formaldehyde are conducted. It is attempted to find a cycle-resolved measure of the chemical effect of PI. However, not any noticeable correlation between the measured formaldehyde content of the charge and main combustion phasing is observed. This can be explained by the findings that the exhaust gas of the prior cycle and potentially pool fire contribute significantly to the measured formaldehyde content of the charge. Thus, formaldehyde is not a good indicator of the chemical effect of PI on main combustion under these conditions.
