Search Results

Gasoline and operating factors affecting the vapor locking tendency of European cars have been studied. Three different test procedures were employed: one, similar to the CRC procedure, was shown by a consumer reaction test to induce vapor lock at ambient temperatures at least 11 F lower than required in normal service. The variation in vapor locking tendency for cars of a given make and model had a standard deviation of 0.61 lb RVP. The tendency of a gasoline to give vapor lock was predicted better by a combination of RVP and front end slope than by some other volatility expressions.

A method is proposed to quantify the activity of a detergent at a selected engine operating condition by measuring the amount of additive required to reduce nozzle fouling to zero in a well-defined engine test procedure. To determine this critical detergent concentration, a series of engine tests is performed using increasing levels of additive, until a zero level of fouling is achieved. These results are then used to construct a detergent concentration response map, defining unambiguously the additive activity and allowing quantitative comparisons of different detergents or additive packages. Concentration response plots are presented from both bench engine test series and vehicle road trials which demonstrate the wide variability in the amount of different detergents required to achieve comparable deposit control. From these maps the performance of the detergents may be compared in terms of keep-clean and clean-up activity for the range of engines and operating conditions.

Environmentally adapted diesel fuels defined by the Swedish Government contain extremely low levels of sulphur and have limited aromatics contents. Road trials and pump durability tests of these fuels revealed unacceptable wear in injection pumps due to low lubricity. Additive solutions were identified using bench tests and then proven in field trials. Market experience has substantiated the findings that fuels using the chosen additive give fully satisfactory performance. This paper illustrates how practical solutions to lubricity questions can be found, and is applicable wherever specifications demand fuels requiring a high degree of hydroprocessing.

Vehicle-based studies have shown that a reduction in the aromatic content of gasoline fuels can result in increased NOx emissions from catalyst-equipped vehicles. A study with simulated exhaust gas has shown that light paraffins, especially methane, are unreactive and cause substantial breakthrough of unreacted NO over the catalyst. However, unsaturated exhaust components including aromatics are effective reactants and play a large part in converting NO over the catalyst. Engine tests have shown that methane is predominantly produced by fuel paraffins and olefins, but hardly at all by aromatics. Thus it appears that methane generated during combustion of low aromatics fuels may be the cause, wholly or in part, of the poor NO conversion efficiency observed when catalyst-equipped cars are operated on such fuels. However, it cannot be excluded that low aromatics fuels are associated with increased air-to-fuel ratio which will also contribute to poor NO performance.

A piston ring-pack lubrication model has been developed which takes into account both lubricant viscosity/temperature and viscosity/shear rate variations. In addition, lubricant starvation of the upper piston rings, due to restriction of the oil supply by the lower rings, has been included. Inputs to the model include piston ring profiles (measured using Talysurf profilometry) and gas pressure distributions throughout the ring-pack. The latter were calculated using the (known) combustion chamber pressure diagram at the relevant engine operating conditions. The model was validated by comparing predicted oil film thicknesses with those measured using a laser-induced fluorescence technique on a Caterpillar-1Y73 single-cylinder diesel engine. The engine was run at a range of speeds with two different, fully formulated, multigrade lubricants, and the oil film thickness under each of the piston rings was measured.

Total aromatics have no influence on particulate emissions. This is an unexpected finding from a comprehensive programme to determine the influence of diesel fuel properties on heavy-duty particulate emissions. 30 fuels and five engines representing a variety of manufacture/technology were tested. To reveal causative influences, key fuel properties were intentionally uncorrelated and had a wide range of values. Engine sensitivities to fuel quality were found to differ considerably. Properties that most influenced emissions were sulphur content, density and cetane number. Some engines were totally insensitive to polyaromatics but in others, small influences are possible. The emissions benefits of specific fuel property changes are quantified.

Extensive tests have been carried out with European and Japanese cars to investigate the occurrence of vapour lock and related problems under hot-weather conditions. Different criteria, based on gasoline inspection properties, were evaluated for their ability to control hot-fuel handling performance. Considerable market experience has been accumulated with the flexible control, RVP (mbar) + 7E70, which is preferred by the authors to the vapour/liquid ratio type of control because it is far easier to use in routine refinery applications and control. However, the two types of control are equivalent in their ability to predict the performance of a wide range of gasolines, including those which may result from reductions in lead content. In their tolerance to gasoline volatility, the European and Japanese cars are very similar to the catalyst-equipped California models tested by the CRC in 1975.

The role and required function of a diesel fuel detergent are discussed and related to the process of nozzle deposit formation. A bench engine test method is described which uses an air-flow technique to measure the fouling produced in IDI nozzles and hence quantify the effectiveness of detergents in the engine test. Data are presented demonstrating the discrimination in detergent performance which may be obtained by applying this method at different operating conditions. Results of ECE 15.04 emissions tests are presented from vehicles having nozzle fouling levels ranging between 1-70%. These emission levels are plotted versus nozzle fouling. From these data, it is concluded that reducing fouling produces a systematic reduction in unburnt hydrocarbon emissions. An optimum fouling level of 15-40% is identified to minimise particulate emissions from this engine.

A set of model fuels has been designed, using the Major-Component Fuel approach, to represent a range of gasoline mid-range and back-end volatilities. The thermo-physical properties of the model fuels have been used, together with a simple model of inlet system, to calculate liquid-vapour mass fractions in the inlet system, and the composition of the inlet port wall film. This has enabled the effects of gasoline volatility, speed, load and inlet port wall temperature to be studied systematically. The results indicate that, in cold start, only some 20-30% of the injected fuel is vapourised in the inlet port, leading to an accumulation of liquid fuel in the inlet port wall film reservoir. As the engine warms up, the mass of fuel in the reservoir decreases, and its composition changes, becoming progressively richer in heavy end species. Mid-range volatility affects the cold start behaviour, whilst back-end volatility affects the approach to fully-warmed up operation.

Six full range gasolines were tested in two engines (one with a catalyst) operated at 4 steady states. Engine-out regulated emissions responded to equivalence ratio, Φ, in the accepted manner. For both CO and NOx, there was a characteristic, single emissions response to changes in Φ. Changing fuel composition will primarily alter the production of these emissions by modifying the stoichiometric air/fuel ratio, projecting engine operation onto another part of the Φ response curve. These Φ effects, which are independent of engine design, also determine how operating conditions affect engine-out CO and NOx. Speciated hydrocarbon measurements at engine-out and tail-pipe confirm results seen in previous test-cycle based programmes.

Methods of identifying cars responsive to TML, and the connection between a car’s response to TML and its severity are examined. Fuel segregation in the inlet manifold is studied with the object of improving laboratory bench procedures. Laboratory results are supported by customer reaction trials, the techniques of which are fully described. The optimum scavenger combination for use with TML is shown to be 1.0 T ethylene dibromide plus 0.2 T tritolyl phosphate.

A “CARS” System using a modeless dye laser has been extensively calibrated and shown to give average temperatures of acceptably good accuracy. It has been used to measure temperatures in the end-gas of a single-cylinder E6 engine under knocking conditions using propane, commercial butane, iso-octane and a primary reference fuel made up of 90% iso-octane and 10% n-heptane by volume. These measurements show that there is significant heating of the end-gas because of pre-flame chemical reactions for all the fuels except propane. Propane has to be compressed to a much higher pressure compared to the other fuels studied in order to make it knock. At a given engine operating condition, there is significant cycle-by-cycle variation in both combustion and knock.

Combustion chamber deposit (CCD) formation in SI engines is a complex phenomenon which is dependent on a number of fuel and engine parameters. A mathematical model has been developed, based upon a previously proposed mechanism of CCD formation, which describes the physical and chemical processes controlling the growth of deposits in SI combustion chambers. The model allows deposit thickness to be predicted as a function of time, taking into account gasoline composition and factors influenced by engine operating conditions. Piston top deposit thicknesses predicted by the model for 38 unadditivated fuels show a strong correlation with data from three different bench engine tests. The model offers the possibility of predicting the amount of CCD produced by unadditivated gasolines for a range of engine designs, operating conditions and test durations.

A series of engine dynamometer tests was carried out with 100% ethyl ester of soya oil as fuel and six different diesel engine lubricants. In each case the lubricant became contaminated by unburnt fuel during the tests with measured dilution rates of up to 0.2% of the fuel throughput. The lubricant/fuel mixture eventually underwent degradation to such an extent that phase separation occurred. The tests were terminated when the lubricant lost all dispersancy, as evaluated by a blotter-spot teat. Used oil analysis revealed that rapid oxidation of some of the fatty acid ester components of the fuel diluent had occurred in the later stages of the tests. At the high levels of fuel dilution recorded in these tests there was little difference between the performances of the six lubricants, despite their differing performance categories.

Principal formulation effects on oil-fluoroelastomer compatibility in European specification tests are reviewed, and detailed surface analysis of elastomer samples after exposure to oil reported. The mechanism of fluoroelastomer degradation has been investigated and found to be based on fluorine depletion of a thin surface layer by dispersant additives. Compatibility is determined by both the oil and the detailed composition of the elastomer. Correlations with field performance have not been published for current European elastomer tests. Failing oils have demonstrated satisfactory performance in field trials. Oil preageing affects seal compatibility, but probably differently for gasoline and diesel engine lubricants.

An experimental programme has been carried out to quantify the influence of Combustion Chamber Deposit (CCD) removal on vehicle acceleration performance, fuel consumption and tailpipe emissions in several modern European car models. Vehicles were performance and emissions tested dirty', following accumulation of 16,000 kilometres (10,000 miles) with a light duty cycle, then ‘clean’, following removal of CCDs. This scheme was repeated for one model using a heavy duty driving cycle. Additional tests were carried out on three vehicle models equipped with knock-sensors for which ignition timing was monitored. CCDs reduced fuel consumption relative to the clean engine, in amounts dependent on vehicle model. CCDs had only small, detrimental effects on acceleration performance and power. They generally (but not always) increased NOx emissions and had variable and usually small effects on HC and CO emissions.

Net heat release rates and knock characteristics were derived from in-cylinder pressures for different fuels in a single-cylinder engine; the effect of an ashless antiknock, N-methyl aniline (NMA) was also studied. The maximum net heat release rate (MHRR) resulting from the final high-temperature chemistry determines the knock intensity. Paraffinic fuels have similar knock intensities at comparable knock occurrence frequencies. Aromatic fuels have significantly lower MHRRs and give much lower mean knock intensities for a given knock occurrence frequency compared to paraffinic fuels. Adding NMA to a paraffinic fuel increases the spark advance required to get a chosen frequency of knock occurrence as it increases the octane number of the fuel but has little effect on MHRR and hence knock intensity.

A unique time-resolved Fourier Transform InfraRed (FTIR) spectrometry technique has been developed to obtain full mid-IR lubricant spectra directly from the ring-pack region of a firing, single cylinder, diesel engine. Initial studies of the detailed spectra show a growth of oxidation products, as indicated by a strong carbonyl absorption peak, observed to increase with load close to the top ring location, for both power and exhaust strokes. Similarly, the formation of alcohol, ketone, aldehyde and carboxylate oxidation products is accessible. Thus it is possible to gauge gross changes to lubricant composition as a function of spatial location through the ring-pack, engine stroke and the severity of engine operation.

As a result of increasing concerns over air quality, environmental legislation has led to more stringent emissions limits for diesel engines and vehicles. This has affected both engine manufactures and fuel suppliers. Whereas in the US, only the fuel requirements for heavy-duty diesel engines are of key interest, in Europe light-duty diesel applications are also important since diesel-powered passenger vehicles are accepted by customers and their market penetration has increased rapidly. This paper gives an update of Shell's ongoing research on correlations between diesel fuel quality and particulate emissions in both heavy- and light-duty applications. In heavy-duty testing (both steady-state and transient), sulphur is the dominant fuel property affecting particulate emissions. After sulphur correction, fuel effects are small and can best be described by a combination of cetane number and density.

Mid-infrared reflection-absorption spectroscopy, has been applied for the first time to the measurement of lubricant degradation products in the ring pack of a firing single-cylinder, IDI diesel. An IR-transmitting window, mounted in the cylinder wall, enables illumination of the moving piston by a broadband IR source located on the engine exterior. Light reflected from the piston is analysed in three wavebands to measure carbonyl oxidation products and oil volumes. Intra-cycle observations reveal differences in the apparent extent of lubricant oxidation between strokes and at different spatial locations in the ring pack. The data are interpreted in terms of a non-homogeneous sample.