Search Results

Previous work has shown that it may be advantageous to use gasoline type fuels with long ignition delays compared to today's diesel fuels in compression ignition engines. In the present work we investigate if high volatility is also needed along with low cetane (high octane) to get more premixed combustion leading to low NO and smoke. A single-cylinder light-duty compression ignition engine is run on four fuels in the diesel boiling range and three fuels in the gasoline boiling range. The lowest cetane diesel boiling range fuel (DCN = 22) also has very high aromatic content (75%vol) but the engine can be run on this to give very low NO (≺ 0.4 g/kWh) and smoke (FSN ≺ 0.1), e.g,. at 4 bar and 10 bar IMEP at 2000 RPM like the gasoline fuels but unlike the diesel fuels with DCNs of 40 and 56. If the combustion phasing and delay are matched for any two fuels at a given operating condition, their emissions behavior is also matched regardless of the differences in volatility and composition.

Experimental results from three different single cylinder engines using the same fuel at different operating conditions and at different stages of build-up of combustion chamber deposits (CCD) are reported. Two of these engines have the same pent-roof, four-valve head but different compression ratios while the other has two valves and a much slower burn-rate. Increasing the compression ratio increases HC and NOx emissions correlate very well with peak engine pressure. The effects of CCD on combustion rate, power and emissions depend greatly on engine design and also on operating conditions.

A new field problem associated with flakes of combustion chamber deposit (CCD) getting trapped on the exhaust valve seat has been reported by several car manufacturers in Europe. This causes difficulties in start-up and poor driveability. A road test procedure that is reasonably quick and sensitive to fuel changes has been developed to study the deposit flaking problem. The flaking of the deposits is believed to be caused by water - either generated by combustion or existing in the ambient air as water vapour - condensing on the deposits. Water is much more effective than fuel in causing deposit flaking. A way of quantifying the deposit flaking tendency has been defined and its repeatability established based on twenty-nine tests using two different cars and different fuels and additives. There are large differences between base fuels in terms of CCD flaking.

A study has been undertaken with a single-cylinder engine, based on the Mitsubishi GDi combustion system, that has the option of either port injection or direct injection. Tests have been undertaken with pure fuel components (methane, iso-octane, toluene and methanol), and a representative gasoline that has also been tested with the addition of 10% methanol and 10% ethanol. The volumetric efficiency depends both on the fuel and its time and place of injection. For stoichiometric operation with unleaded gasoline, changing from port injection to direct injection led to a 9% increase in volumetric efficiency, which was improved by a further 3% when 10% methanol was blended with the gasoline. The improvements in volumetric efficiency will be used to quantify the extent of charge cooling by fuel evaporation, and these will be compared with predictions assuming the maximum possible level of fuel evaporation.

Oxides of nitrogen (NOx) and smoke can be simultaneously reduced in compression ignition engines by getting combustion to occur at low temperatures and by delaying the heat release till after the fuel and air have been sufficiently mixed. One of the ways to obtain such combustion in modern engines using common-rail direct injection is to inject the fuel near top dead centre with high levels of exhaust gas recirculation (EGR) - Nissan MK style combustion. In this work we study the effect of fuel auto-ignition quality, using four fuels ranging from diesel to gasoline, on such combustion at two inlet pressures and different EGR levels. The experiments are done in a 2 litre single-cylinder engine with a compression ratio of 14 at an engine speed of 1200 RPM. The engine can be easily run on gasoline with a single injection near TDC, even though it cannot be run with very early injection, in the HCCI mode.

Combustion chamber deposits (CCD) increase more slowly with time and take longer to stabilise in a given engine test with a base fuel compared to the same base fuel with an additive package. In a short fixed-duration test the additive might appear to increase CCD levels significantly even though the stabilised CCD levels for the two fuels are not too different. When the same additive package is used in different base fuels, the fuels are ranked in the same order with or without the additive. However, the incremental increase in CCD because of the additive is more marked the cleaner the base fuel.

A four-cylinder engine with a slice between the head and the block carrying instrumented plugs has been used to study the growth of combustion chamber deposits and knock. Deposit thicknesses vary substantially at different locations, the thickness generally being greatest at the coolest surfaces. If a dirty engine is run on a low-boiling-point fuel such as a primary reference fuel, deposits are removed and octane requirement is reduced rapidly. Of the head deposits, those in the cooler squish region where the end gas is likely to be situated affect knock more than the deposits in the hotter regions. Different fuel additives have different effects on deposits in different areas. For instance, an additive might cause a substantial increase in deposit thickness in the hotter areas and a slight increase in total deposit weight but can control deposits in the cooler squish regions and so reduce octane requirement increase (ORI) compared to the base fuel alone.

In previous work it has been shown that the autoignition quality of a fuel at a given operating condition can be described by its Octane Index, OI = (1-K)RON - KMON; the larger the OI, the more the resistance to autoignition. Here RON and MON are, respectively, the Research and Motor Octane numbers of the fuel and K is a constant depending only on the pressure and temperature history of the fuel / air mixture in the engine prior to autoignition. The value of K is empirically established at a given operating condition by ranking fuels of different RON and MON and of different chemical composition for their ease of autoignition. Another important parameter at a given operating condition is OI0, the Octane Index of the fuel for which heat release is centred at TDC. In previous work K and OI0 were measured at different operating conditions and were related empirically to pressure and temperature of the mixture before autoignition and to engine speed and mixture strength.

The auto-ignition or anti-knock quality of a practical fuel is defined by the Octane Index, OI = (1-K)RON + KMON where RON and MON are the Research and Motor Octane numbers and K is a constant depending only on the pressure and temperature variation in the engine. K decreases as the compression temperature in the unburnt gas at a given pressure in the engine decreases and can be negative if this temperature is lower than in the RON test. As spark ignition (SI) engine designers seek higher efficiency knock becomes more likely. Moreover such initiatives - direct injection, higher compression ratios, downsizing and turbocharging - will reduce the unburnt gas temperature for a given pressure and push the value of K downwards. In Europe there is evidence of a monotonic decrease in the average K value from 1987 to 1992. In 37 different Japanese and European cars (34 models) equipped with knock sensors that have been tested K has been found to be negative in most cases.

Combustion and pressure development in spark ignition engines are marked by cycle to cycle variations which are especially severe if the mixture is lean. The variations in indicated mean effective pressure (IMEP) that arise from this could be sufficiently severe to cause problems in certain engine operating regimes even in engines which run smoothly at steady operating conditions. Cyclic variations in IMEP could be effectively reduced at source by reducing cyclic variations in combustion. These are known to originate during the initial stage of combustion which can be influenced by the ignition process. Of the various enhanced ignition devices, variants of the spark ignition system seem to be of most practical interest for automobile applications in the immediate future. In the paper, we review the literature on spark ignition, the nature of the spark discharge process and the attempts to improve stability of engine operation through changes in spark ignition.