Performance and Emissions of a Turbocharged Spark Ignition Engine Fuelled with CNG and CNG/Hydrogen Blends

An experimental investigation was performed on a turbocharged spark-ignition 4-cylinder production engine fuelled with natural gas and with two blends of natural gas and hydrogen (15% and 25% in volume of H₂). The engine was purposely designed to give optimal performance when running on CNG. The first part of the experimental campaign was carried out at MBT timing under stoichiometric conditions: load sweeps at constant engine speed and speed sweeps at constant load were performed. Afterwards, spark advance sweeps and relative air/fuel ratio sweeps were acquired at constant engine speed and load. The three fuels were compared in terms of performance (fuel conversion efficiency, brake specific fuel consumption, brake specific energy consumption and indicated mean effective pressure) and brake specific emissions (THC, NO_x, CO). The pressure trace was acquired in the four cylinders to perform a cycle-by-cycle and cylinder-by-cylinder analysis of the peak firing pressure and the indicated mean effective pressure as well as of the main combustion parameters.

As far as MBT timing stoichiometric operations are concerned, the addition of hydrogen determines a reduction in the spark advance. The higher energy content of hydrogen with respect to methane determines lower brake specific fuel consumption values. Conversely, similar values were found for the brake specific energy consumption, which is intimately related to the fuel conversion efficiency. The addition of H₂ produces an appreciable reduction in THC which overcomes the mere reduction of HC in the blend. A small improvement can also be found for CO emissions whereas NO_x turned out to be slightly affected by the H₂ content in the fuel.

The air/fuel ratio in the relative air/fuel ratio sweeps was varied in the rich and lean field up to the lean operation limit. The addition of hydrogen determined a marked reduction of the coefficient of variations in the lean field, thus allowing an extension of the lean operation limit with respect to CNG operation. The addition of hydrogen still determined a considerable THC reduction which turned out to be even more evident as one moved to the lean field.

Regardless of the considered fuel, a significant cylinder-to-cylinder variation emerged, mainly to be attributed to the cylinder spatial distribution, which in turn affects the effectiveness of the cooling system and the uniform air/fuel mixture distribution.