资源描述
Toward Higher Speeds and Outputs From the Small Diesel EngineRicardo & Co.Engineers(1927) Ltd.(England)THE AUTHORS company has long been concerned with the development of the small, high-speed diesel engine, and is particularly associated with combustion systems for this type of engine. Although such engines are not common in the North American continent, production and use in Europe and Japan is considerable, totaling several million units. These are, typically, naturally aspirated 4-cyl engines of 25-35 in3 (400-600cm3) displacement per cylinder, operating up to speeds of 4000-5000 rpm, with a limiting piston speed of about 2400ft/min(12m/s). In discussion with the U.S.Army Tank-Automotive Command (USATAC) at , Mich.it was proposed that the military requirement of high power from a small lightweight package could be achieved by exploiting higher speeds than hitherto, rather than the application of increased levels of turbocharger alone, and this led to the formulation of a research program to study combustion and breathing problems under such conditions. This paper describes the work carried out to date, which has involved the design, manufacture, and preliminary test work on special single cylinder engine. THE PROJECT The project specifications finally laid down by USATAC can be summarized as follows:1. Design, procure, build, and test a single-cylinder engine of 3-1/2 in (88.9mm) bore and stroke, to operate at the highest possible speed, but certainly above 5000 rpm. Simulation of turbocharged conditions to be achieved using a separate air supply.2. To develop the single-cylinder test engine to achieve performance targets such that a 4-cyl version for military duties could produce 1 bhp/in33) displacement with a target dry weight of about 3.5 lb/bhp (2.13kg/kW).3. The design not to be influenced by conventional practices, with the aim of minimizing mechanical and thermal stresses.4. Operation on CITE-R fuel (MIL-F-46005A (MR) to be the primary requirement. Initially, fuels down to aviation gasoline were to be investigated, but this latter requirement was subsequently relaxed.5. Lubricating oils to the MIL-L-2104B specification to be used if at all possible.6. The final phase of the project to include a design study for a 4-cyl military engine, embodying the lessons learned on the single-cylinder test unit.7. Starting, idling, and light-load operation of the multicylinder engine must not be compromised.PRELIMINARY DESIGN CONSIDERATIONS A simple examination of the cylinder size and power output target rapidly showed the limitations that the maximum engine speed would have on performance (Table 1). Starting from the minimum speed specified of 5000 rpm, it is clear that speeds of 6000 rpm and above entail piston speeds equal to those of racing gasoline engines. While the reductions in bmep through use of high speeds are significant, the increases in fmep (estimated from past results obtained at the authors company, much of which has been summarized in Ref.1) give very little return in reduced imep. Naturally aspirated automotive diesel engines working to the strict smoke limits of a few years hence can only operate up to about 145lb/in2 (1000kPa) imep; hence it was clear that some measure of turbocharging would be required. A further penalty of high speed and high engine friction is in fuel consumption, and Table 1 makes clear how the bsfc would worsen rapidly to levels no better than a gasoline engine, so losing one of the major advantages of the compression ignition cycle. In these circumstances,it was decided to limit the speed of the research engine to 6000 rpm. The major performance problems involved in the design of an engine to meet these requirements might be summarized as follows: ENGINE BREATHING-Previous experience on small high-speed diesels had shown that the major limitation on imep at high piston speeds is the breathing of the engine (2).Hence, valves of sufficient flow area had to be provided to allow efficient operation up to 3500 ft/min (17.8m/s) piston speed, some 50% higher than levels normally employed in diesel engines. This would certainly require departures from conventional cylinder head arrangements, involving inclined multiple-valve designs (Table 2);turbocharged operation brings a slight bonus in that the higher inlet air temperatures minimize pressure losses and reduce volumetric efficiency changes.In addition, possible turbocharger matching requirements had to be borne in mind. While, for automotive engines, torque backup requirements normally favor minimizing the available boost at the rated speed, so that a large exhaust valve area is not mandatory, in this case the very short absolute exhaust gas release periods suggested that the exhaust mean gas velocities should be kept low, and exhaust valve area about equal to that of the inlet. Also requiring consideration was the question of valve timings. For the inlet, high speeds are normally associated with a late closing point, yet in the case of a diesel, and with closing points later than about 45 deg abdc, there would be a progressive sacrifice in starting ability, as well as some loss of low-speed performance, which would further impair the natural torque backup characteristics of the engine. For the exhaust, the turbocharger matching requirement again dictates an early release of the gases on the expansion stroke, and timings later than about 60 deg abdc do not show to advantage at high speeds. While a long overlap period could contribute to reduction of exhaust gas and exhaust system component temperatures, such gains would be minimal at high speeds due to the very low quantity of scavenge air which might be passed relative to the trapped flow, and the mechanical problems of obtaining the piston/valve clearance would place a severe penalty on the combustion system. COMBUSTION PROBLEMS-A fundamental problem likely to affect the engine at the speeds contemplated was the likely duration of the ignition delay period. Ignition delay is a function of engine speed, compression conditions, and injection timing for a fuel of particular ignition delay is a function of engine speed, compression conditions, and injection timing for a fuel of particular ignition quality at normal running conditions (3),if factors related to the particular combustion chamber configuration in use are considered as being of second order. CITE-R fuel has a minimum specified cetane rating of 37,but the published data on engine delay using this fuel covered only low-speed conditions, and were not of direct use in predicting results at 6000 rpm. However, consideration of these available data, together with the known performance of small high-speed engines operating up to 5000 rpm on gas oil (55 cetane), led to the estimates shown in Fig.1,for the lowest compression ratio which would allow acceptable starting (higher ratios would give excessive heat losses and maximum cylinder pressures).These suggested that unaided or true compression ignition operation at 6000 rpm was feasible on CITE fuel, although the light-load condition would require inlet manifold air temperatures to be maintained significantly above ambient-not an impossible requirement for a turbocharged engine. Injection periods being controlled by the injection system will depend on the latters type, but for practical reasons there could be no possibility of developing new systems for the project, to replace the conventional jerk pump arrangement. With a fixed orifice area nozzle, there would be considerable problems in passing the required full load quantity of up to about 60 mm3/injection at 6000 rpm at the required rate, yet obtaining satisfactory characteristics for idling, the turndown ratio being about 11:1.The effect of an extended injection period on combustion at the high speeds required could be very severe on a direct injection (DI) combustion system, where use of a fixed orifice nozzle would be inevitable. In addition to this problem the major difficulty of the DI was seen as the high mechanical loading, accentuated by the higher smoke-limited fuel/air ratios (A/F) requiring higher boosts to achieve the target rating. While Ricardos earlier research work had shown that DI systems could be made to operate up to 4500 rpm naturally aspirated, on balance (see Table 3) the Comet swirl chamber system, developed over many years for the small high-speed commercial engine, was considered to offer greater potential for this particular application. The major problem foreseen was high thermal loading, although the unaided starting and the full multifuel capabilities were also less satisfactory than those of the DI: however, with built-in aids such as could be applied to the multicylinder engine, and with a restriction to CITE fuel, these latter were not considered to be too serious.At the start of the project, then, some consideration was given to the DI as an alternative, and in addition to test running under boosted conditions of an existing 4000 rpm single-cylinder research unit, designs were completed for a DI version of the test engine. Since that date, the increased pressure of noise, smoke, and particularly exhaust emissions legislation, has increasingly favored the divided chamber system, and test work on the DI version is not now likely to take place.ENGINE FRICTION-Comparing the proposed multicylinder high-speed turbocharged engine with a conventional commercial engine of the same cylinder size and number, it was clear that the former would have a significantly higher fmep through the use of higher rotational and mean piston speeds. As already made clear in Table 1,this would pose serious problems in relation to the attainment of both the target output and an acceptable fuel consumption.Fig.2 shows how using a typical commercial engine fmep/speed curve from Ref.1, the estimated fmep of the high-speed multicylinder engine was obtained. In this estimate, some increase in the mechanical friction of the basic engine structure was assumed, since the turbocharged condition would increase cylinder pressures and require larger bearings to give acceptable reliability. In addition, inlet and exhaust pumping losses could add materially to the high-speed fmep, unless acceptable valve sizes could be maintained. By gasoline standards,then,the mechanical efficiency of the unit would be poor,but experience had shown that although attention to detail throughout the design could yield gains,these low levels were implicit in the project specification.SINGLE-CYLINDER TEST ENGINE Based on the considerations outlined, the definitive single-cylinder test engine was designed, the boundary operating conditions for the engine being: bore and stroke,3-1/2 in 3-1/2 in (88.9mm88.9mm);normal full-load speed range,3000-6000 rpm; and maximum cylinder pressure,2500 lb/in2(17.3Mpa). While the cylinder pressure limit may seem high by conventional standards, past experience has shown the dangers of designing such engines to low limits, and thus inflicting unforeseen limitations on the test program. In fact, originally, with possible work on a DI version in mind, a limit of 3000 lb/in2(20.7Mpa) was set, but as noted, this limit was later reduced for the Comet version. The Comet swirl chamber engine layout is illustrated in Figs.3-5, and the complete engine shown in Fig.6.Of the major components, the following may be said: CRANKCASE-The crankcase and rear-mounted timing case and cover are in gray flake graphite iron to BS 1452:1961 Grade 14,spigoted or doweled and bolted together. The crankcase design was adopted from that of the Ricardo E/6 variable compression ratio gasoline engine, which results in the presence of the front chamber of the crankcase unit, where the E/6 timing drive was situated. Three main bearings are use, all of lead-bronze bushing type, the center bearing being the thrust bearing. The rear bearing acts only as a steady bearing to the otherwise long extension of the crankshaft, the clearance being adjusted so that it cannot take the firing load off the center bearing. CRANKSHAFT-The crankshaft is a one-piece forging in nitriding steel to BS 970:1955 En 40c.The balance weights are integral, and balance only the rotating loads, since primary and secondary balancer shafts are fitted to the engine. All journal and pin surfaces are nitrided, the diameters of the three journals being 3,3,and 2-3/8 in (76.2,76.2,and 60.4 mm),respectively, from front to rear, and of the pin 2-5/8 in (66.6mm). CONNECTING ROD-To obtain the better material properties associated with a forging without the expense of special dies, a search was made of commercial engines, and the connecting rod of the Ford 2700 series diesel engine finally selected as the most appropriate. While satisfactory big-end bearing loadings were achieved at the 3000 lb/in2 maximum cylinder pressures, the little-end design was considered inadequate, and the decision made to use this rod only for the Comet swirl chamber version of the engine, at a pressure limit of 2500 lb/in2. The bearings are as used on the 2700 series engine, that is ,15% reticular tin/aluminium half liners, the little-end bushing being a wrapped lead-bronze item. In addition to careful checking and polishing of the rod, the little end is reduced in width, the better to distribute the firing and inertia loads between the piston pin bosses and the little-end bushing. A higher torque than standard is used on the big-end setscrews, to prevent the cap lifting off due to the inertia forces at tdc exhaust, at 6000 rpm. Computer calculations carried out by the bearing suppliers, The Glacier Metal Co.Ltd., showed that the proposed bearing arrangements were acceptable, although the big end in particular has to accept very arduous conditions at high speeds due to the great inertia of the relatively massive connecting rod (Fig.7).The importance of correct form for both the pin and the bearing under these conditions cannot be overstressed; this apart, the only problem that occurred was rapid cavitation attack in the top (loaded) half liner with the original clearance. The cause of this is evident in Fig.7,and a reduction in clearance to 0.0022 in (56m) cured this trouble. PISTON AND WRIST PIN-The piston is a one-piece sand casting in 13% silicon aluminum alloy to BS 1490:1970 LM13WP,with the shallow trench and twin recesses of the Comet combustion system formed in one face of the angled (pent-roof) crown surface. Two compression rings are used, the top being a plain barrel-faced ring and the second a taper-faced internally stepped (twisted) ring; the slotted oil-control ring is of the conformable type. Rings are supplied copper-plated on the rubbing faces to assist in bedding in, but are not chrome-plated, since this facing is applied to the liner. The piston was designed deliberately of relatively great height, since it was feared that the very high (by diesel standards) piston speeds together with boosted operation would create difficulties in obtaining acceptable piston, ring, and liner conditions, and it was not thought desirable to accentuate problems more than was necessary. However, relatively little trouble has been experienced with the ring pack. Piston cooling and little-end lubrication is via an oil spray from a fixed jet located in the crankcase. This method was selected to avoid grooving the big-end bearing liner, which would have reduced its capacity. Two piston designs were developed, a tray-cooled arrangement and the soluble core design shown in the figures. To obtain acceptable cooling of the ring belt with the tray-cooled design, the struts transmitting gas loads from the crown to the wrist pin bosses were thinned as far as was thought practicable, but this arrangement was found to allow excessive distortion. The soluble core design has given excellent service to date. The wrist pin is of case-hardened steel,1-3/8 in (34.9mm) in diameter: CYLINDER LINER AND WATER JACKET-The high rates of local heat transfer associated with the use of a swirl chamber combustion system, together with the 2500 lb/in2 maximum cylinder pressure limit, led to design difficulties with the wet-type cylinder liner, since calculations showed that a conventional iron liner of thickness adequate to withstand the gas loads would give excessive surface temperatures for acceptable lubrication at the top ring reversal point. The solution adopted was to use a steel liner, with the bore given the necessary surface finish before being plated with hard chrome to a thickness of 0.0015 in (38m) by the Chromard process. Toward the top, the liner is thinned to provide the necessary temperature control, while the greater thickness lower down enhances rigidity to combat water-side attack. The liner is flanged at the top and seats on the cylindrical mild steel water jacket itself seating on top of the crankcase; radial location is provided by the liner spigoting in the crankcase, a water seal being obtained in the normal way by rubber O-rings. CYLINDER HEAD ASSEMBLY-The cylinder head, with its associated cambox, is the most complex single assembly of the engine, and presented considerable design problems. The most pressing of these centered on the provision of adequate valves and ports-the difficulties here may be appreciated when it is realized that ports suitable for a conventional engine of 5-1/2 in bore had to be provided on a 3-1/2 in bore-together with adequate cooling for the very high rating of 5.7 ihp/in2 (0.66 indicated kW/cm2) of piston area, this with a swirl chamber comber combustion system. The position of the Comet swirl chamber at the edge of the bore does not render the use of four valves very attractive, and although this and other possible layouts were examined, a 3-valve arrangement was finally adopted. A pent-roof head surface was necessary (Table 2), partly to obtain the necessary valve area but primarily to prevent excessive congestion higher up in the head. It is normally preferable to pair the exhaust valves to reduce their individual size and use a single large valve, but the opposite layout was finally chosen, as shown in Fig.8, since the paired valves had to lie in the center of the head and the intense heating of the head from the long port duct if this latter were the exhaust was considered unacceptable. Asymmetrical chamber/valye layouts were also investigated but rejected as offering no real advantages and being incompatible with multicylinder requirements. The so-called externally inserted form of the Comet chamber was adopted to minimize the space occupied by the hot plug forming the lower portion of the chamber. To cool the resulting four bridges ins the lower deck of the head, between the chamber and the inlet valves, and the inlet and exhaust valves, drillings were provided, giving an accurately controllable metal thickness between the hot gases and the coolant, and a clean surface on the coolant side. The swirl chamber hot plug carrying the throat, and made from Nimonic 80A alloy by precision casting, is a light fit on its sides as well as locating on a copper gasket on its upper flange, since experience showed that at these high ratings some direct cooling was necessary, unlike commercial engines where an air gap is used to improve warm up after starting. The upper part of the chamber carrying the injector is a spheroidal graphite iron casting. A transducer tapping into the cylinder is provided at the front end of the head. The light alloy head seats directly on the steel liner flange, no gasket being employed, and is clamped by eight suds rooted high in the head and passing vertically downward through the water jacket top flange. No difficulties with gas blow have been experienced, and it has been found possible to operate with a head stud load of only
展开阅读全文