Search Results

The combustion system developed for the General Motors GT-309 regenerative gas turbine is used to illustrate pertinent structural, performance, and exhaust emission considerations when designing for a vehicular gas turbine application. The development of each major component and the performance of the combustion system as a whole are reviewed. The satisfactory performance and durability potential of the GT-309 engine combustion system have been demonstrated by extensive operation in a component test facility and in several test cell and vehicle installed engines. Exhaust emissions of unburned hydrocarbons and carbon monoxide are minimal and are of no concern from an air pollution standpoint. No objectionable exhaust smoking and odor are produced.

During development of the General Motors rotary engine, the lubricant was recognized as important to its success because certain lubricants produced deposits which tended to stick both side and apex seals. Consequently, it was decided to develop a rotary engine-dynamometer test, using a Mazda engine, which could be used for lubricant evaluation. In an investigation using an SE engine oil with which there was rotary engine experience, engine operating variables and engine modifications were studied until the greatest amount of deposits were obtained in 100 h of testing. The most significant engine modifications were: omission of inner side seals, plugging of half the rotor bearing holes, pinning of oil seals, grinding of end and intermediate housings, and using a separate oil reservoir for the metering pump. Using this 100 h test procedure, three engine oils and five automatic transmission fluids were evaluated.

Studies were conducted using spark-ignited reciprocating engines to evaluate ammonia as an alternate fuel for certain military applications. Conventional engines were found to perform poorly on ammonia. Several practical methods for improving engine performance while burning ammonia are described which include increased spark energy, increased compression ratio, engine supercharging, and hydrogen addition to the fuel. Dissociation of ammonia was investigated as a practical means for supplying hydrogen to an engine. The study indicates that satisfactory engine performance can be obtained while burning ammonia. Auxiliary equipment and controls necessary for vehicular use will require development.

Tests of laser ignition are conducted in a combustion bomb. A range of fuels is investigated comprising isooctane, cyclohexane, n-heptane, n-hexane, clear indolene, and No. 1 diesel fuel. The ignition characteristics of laser-induced sparks are compared with sparks generated with a spark plug for different air/fuel ratios. The power density required to produce laser induced sparks is investigated. Although laser ignition appears to be impractical as an ignition device because of its low efficiency and high cost, it presents some interesting possibilities compared to the standard spark plug in that the laser spark is electrodeless and can be positioned anywhere inside the combustion chamber. Its primary use appears to be as a research tool.

Secondary air scheduling and average delivery rate have a great influence on the performance (carbon monoxide and hydrocarbon cleanup) of rich thermal manifold reactors. A continuously modulated secondary air system was devised to provide a tailpipe air-fuel ratio that did not change significantly with engine speed or load when a “flat” carburetion calibration was incorporated. This system involved throttling the inlet of the air pump(s) so that the air pump and engine intake pressures were equal. The continuous air modulation system was compared with an unmodulated system and a step-modulated system. The secondary air systems were investigated with both GMR “small volume” cast iron thermal reactors and Du Pont V thermal reactors on modified 350 CID V-8 engines in 1969 Chevrolet passenger vehicles. It was found that thermal reactor performance improved with each increase in control of the secondary air schedule.

The development of a variable valve timing (VVT) camshaft was initiated as a potential means of controlling exhaust emissions from a spark ignition piston engine. This approach was based on the fact that valve overlap influences internal exhaust gas recirculation which in turn affects spark ignition engine emissions and performance. The design, fabrication, bench tests and engine durability tests of a unit incorporating splines to allow the intake cams to move relative to the exhaust cams is discussed. Preliminary test data from a 350 CID (5700 cm3) engine fitted with the VVT camshaft are discussed with regard to durability and emissions.

This paper illustrates by way of two practical examples, namely, transmission gears and crankshafts, how the automotive industry applies basic approaches and methods for achieving fatigue resistant design. Analytic, laboratory, and field studies necessary in the development of these components are briefly outlined.

In-cylinder sampling appears to be the only available means for obtaining detailed information of the diesel combustion process. This information is necessary to understand pollutant formation because of the intimate relationship between formation rates and local cylinder conditions. This paper discusses efforts to (1) examine and improve sampling valve design, (2) evaluate potential effects of the valve and the sampling system on sample composition, (3) find methods to extract useful information from sampling data. Sampling hardware is currently being used to study combustion in engines, but further work is needed to quantify the influence of hardware and procedures on sample composition and to design experiments to provide data containing maximum information.

Paper deals with abnormal combustion initially existing in the Hyprex gasifier. It was found that this was mainly due to a long ignition delay period. To correct this situation, the fuel injection system was redesigned to permit later injection at higher compression pressures with good injection characteristics.

Development of engine-vehicle prototypes for low emissions and optimum fuel control characteristics has been facilitated through use of a computerized emissions test system. Simultaneous on-line sampling of exhaust species concentrations, fuel consumption, spark advance, pressures, and temperatures provides both graphical and computed outputs of several vehicle parameters that are important to development programs. On-line display of vehicle air-fuel ratio is continuously supplied. Either of two federal driving cycles (or any random driving schedule) may be employed. Dynamic calibration, range sensing, and zero-drift correction keep operator interaction and errors to a minimum. Capability for reprocessing, plotting, and/or patching stored data provides increased computational flexibility.

The relationship between surface temperature and exhaust hydrocarbon concentration was explored by installing surface thermocouples at three locations in the combustion chamber of a single-cylinder engine. Coolant temperature, coolant passage surface scale, and ethylene glycol in the coolant affected exhaust hydrocarbon concentration through changes in surface temperature. As power output increased, combustion chamber surface temperature rose, and exhaust hydrocarbon concentration fell. The increase in surface temperature accounted for about 43% of the decrease in hydrocarbon concentration. The reason for the other 57% of the decrease is unknown, but it may have been caused by increasing gas temperatures in the quench zone. Increasing surface temperature by engine modification would be expected to have adverse effects on engine octane requirement, volumetric efficiency, and oil oxidation.

This study describes the effect of spark plug location on NO and HC emissions from a single-cylinder engine with a specially modified combustion chamber. The effects of changes in combustion duration caused either by spark location, dual spark plugs, or charge dilution on NO and HC emissions were also examined. Experiments were run at constant speed, constant load, and mbt spark timing. Nitric oxide emissions were the same with the spark plug located either near the intake or exhaust valve, but were higher with the spark plug midway between the valves or with dual ignition. Hydrocarbon emissions were lowest with the spark plug nearest the exhaust valve and increased with the distance of the spark plug from the exhaust valve. With charge dilution the decrease in NO emission was isolated into a pure dilution effect and a combustion duration effect. The combustion duration effect was minimal at rich mixtures and increased with air-fuel ratio.

A brief review of the testing of automatic transmission fluid for compatibility with seals is presented. The total immersion test used in fluid qualification, while apparently effective in predicting the compatibility of fluids and seals in service, does not correlate well with transmission tests with respect to hardness change of piston seals. The Dip-Cycle Test, developed to overcome this limitation, is a procedure for alternately immersing seal specimens in the test fluid and suspending them in the hot air-fluid vapor atmosphere above the fluid. Correlation of the Dip-Cycle Test with transmission piston seal results is much improved over that with the total immersion test. It is the purpose of this paper to review these developments and to present an improved test procedure (dip cycle test) for evaluating the effect of fluids on transmission piston seal materials.

The exhaust hydrocarbon and nitrogen oxide concentrations of a single-cylinder engine, operating on a 25% (wt.) ethyl alcohol – 75% gasoline fuel, are compared to those operating on gasoline. For comparisons at the same airfuel ratio but lower than 15.3, the addition of ethyl alcohol to gasoline reduces the exhaust hydrocarbon concentrations and increases the nitrogen oxide concentrations. At the same air-fuel ratio but higher than 15.3, the addition of ethyl alcohol reduces both the hydrocarbon and nitrogen oxide concentrations. However, tests with automobiles, operating at the same air-fuel ratio with both fuels, indicate that the addition of ethyl alcohol causes an increase in “surge” and, in some cases, results in a power loss. To overcome these performance problems, the ethyl alcohol-gasoline fuel should be operated at about the same percent theoretical air as gasoline.

THIS PAPER discusses the Stirling thermal enging from four points of view: 1. The ideal, thermodynamic point of view, showing the inherent potentialities of the ideal Stirling cycle in comparison to the basic cycles of other engines. 2. The physical engine and its method of operation with respect to the ideal cycle and the limitations of practical mechanics. 3. Performance data from the first modern Stirling engines ever operated in the United States, evaluating the relationship between the new engine and other more familiar engines of similar sizes. This comparative discussion serves to demonstrate the advantages and disadvantages of the Stirling engine and to indicate its proper place in the 1960 family of prime movers. 4. A look backward into the century of history behind the modern engine pointing out significant milestones in the engine's development.

The advantages and disadvantages of driving accessories from the gasifier section of a free turbine engine are reviewed. Because of the elevating effect of power extraction from the gasifier shaft on turbine inlet temperature in an existing engine, an analysis is made of factors influencing that temperature. Some of the techniques available for avoiding excessive turbine inlet temperature as a result of heavy accessory loads are listed, several of them being illustrated by past vehicular installations.

Convective heat transfer coefficients were determined for thermocouple junctions oriented parallel to the gas flow for Reynolds numbers (based on wire diameter) from 163 to 17,500. Chromel-alumel wires of 0.013-0.051 in. diameter were tested in air and products of natural gas combustion at temperatures of 60, 500, and 1000 F. Transient response was used to determine the heat transfer coefficient, and all data were corrected for variation of metal specific heat and radiant heat transfer. The affect on apparent heat transfer coefficient was determined for variations in junction weld-bead size, junction length, and wire separation. An empirical equation has been derived relating Nusselt number and Reynolds number that fits 92% of the test data within ± 10%.

The use of relatively small catalytic converters containing alumina-supported platinum (Pt) and palladium (Pd) catalysts to control exhaust emissions of hydrocarbons (HC) and carbon monoxide (CO) was investigated in full-scale vehicle tests. Catalytic converters containing 70-80in3 of fresh catalyst were installed at two converter locations on the vehicle. Carburetion was richer than stoichiometric, with air-fuel ratios (A/F) comparable to those proposed for dual-catalyst systems containing an NOx reduction catalyst. The vehicle was equipped with exhaust manifold air injection. Homogeneous thermal reaction in the exhaust manifolds played a significant role in the overall control of HC and CO. Four Pt catalysts, three Pd catalysts, and one Pt-Pd catalyst were prepared and evaluated. Total metal loadings were varied 0.01-0.07 troy oz. Hydrocarbon conversion efficiencies varied 62-82%, measured over the 1975 cold-hot start weighted Federal Test Procedure.

Noise produced by automotive ignition systems can deteriorate the performance of nearby communication systems. An important step toward alleviating this difficulty is to characterize the ignition noise. Measurements have been made of the noise peak amplitude distribution of a number of identically equipped vehicles over a fixed period of time. Both vertical and horizontal polarizations were used, and measurements were made at two frequencies, 145 and 230 MHz. These statistics were then compared to various probability distributions to attempt to characterize the amplitude distribution of the noise. The distributions studied were: the log-normal, the exponential, the Rayleigh, and the Weibull distributions. It was concluded that the best fit was provided by the Weibull distribution. The parameters of the best fitting distribution are primarily a function of the antenna's polarization, with frequency having only a minor effect.