Internal Combustion Engine Vehicles

K. G. DULEEP
Energy and Environmental Analysis, Inc. Arlington, Virginia, United States

1. Introduction 2. Vehicle Energy Efciency 3. Improving Efciency by Reducing Tractive Energy Required 4. Improvement of Engine Efciency 5. Increasing the Efciency of Spark Ignition Engines 6. Increasing the Efciency of Compression Ignition (Diesel) Engines 7. Intake Charge Boosting 8. Alternative Heat Engines

in marine vessels and aircraft and are not discussed here. A small number of vehicles using the Wankel engine have also been sold. Internal combustion enginepowered vehicles typically account for onequarter to one-third of total energy consumption in most countries, and their fuel consumption and fuel efciency are issues of major concern.

1. INTRODUCTION
The on-highway eet of vehicles accounts for over 95% of all vehicles in operation worldwide (which is in excess of a billion vehicles). The remainder is composed of off-highway vehicles, equipment such as forklifts or bulldozers, and motorcycles. Annual sales of on-highway vehicles exceeded 57.6 millions units worldwide in 2002 with 39.5 millions unit classied as passenger cars and 18.1 million units classied as trucks. The distinction between cars and trucks is not always clear (especially for light trucks) but trucks are usually used for cargo hauling or for carrying more than six passengers. Cars span the gross vehicle weight (GVW) range from 1 to 3 tons, while trucks typically span the GVW range of 2 to 40 tons. Off-road and specialized vehicles can be much heavier. The majority of cars and light trucks (under ve tons GVW) are powered by spark ignition engines, while most trucks that weigh more than 5 tons GVW are powered by diesel engines. Since the early 1990s, diesel engines have become more popular for cars and light trucks in the European Union (EU). The diesel engines share in the new car market was over 50% in 2002 in countries such as France and Austria. In contrast, few diesel enginepowered cars and light trucks are sold in North America.

Glossary
engine efciency Amount of energy produced by the engine per unit of fuel energy consumed. fuel economy Vehicle distance traveled per unit volume of fuel consumed. internal combustion engine vehicle Vehicle where primary motive power is derived from an engine that converts fuel energy to work using the air-fuel mixture as the working uid. light duty on highway vehicles Cars and light trucks with a fully loaded weight below 6000 k (13,200 lbs). off-highway vehicle Vehicles designed to operate primarily on unpaved surfaces.

The vast majority of vehicles used in the world are powered by internal combustion engines (ICE). Other forms of propulsion such as electric motors or external combustion steam engines are used in specialized applications that account for a small fraction of the total vehicle eet. Most vehicles are now powered by reciprocating piston engines that use the Otto cycle (also called the spark-ignition engine) or the diesel cycle (also called the compression-ignition engine). Gas turbines are used primarily

Encyclopedia of Energy, Volume 3. r 2004 Elsevier Inc. All rights reserved.

497

498

Internal Combustion Engine Vehicles

2. VEHICLE ENERGY EFFICIENCY

Vehicle energy efciency is generally dened in terms of fuel economy measured in miles per gallon (mpg) or kilometers per liter of fuel. It can also be measured in terms of fuel consumption, which is the inverse of fuel economy, in units of liters per 100 km or gallons per 100 miles. The fuel economy of a vehicle is strongly dependent on the vehicles overall weight, but is also dependent on the efciency of the engine, as well as the matching of the engine characteristics to the vehicles operational requirements. The fuel economy of a particular vehicle is dependent on the load carried, the driving cycle, the ambient temperature, and the characteristics of the road such as its gradient and surface roughness. Hence, the fuel economy of a specic vehicle can vary widely depending on how and where it is used. In most developed countries, on-road light vehicles (cars and light trucks) are certied for emissions and fuel economy by the government. The fuel efciency rating is measured in a laboratory-controlled environment and on a specied driving cycle. In the United States and Canada, for example, light vehicle fuel economy is measured on a city cycle with an average speed of about 20 mph and a highway cycle with an average speed of about 50 mph. All aspects of the fuel economy test, ranging from the ambient temperature to the specication of the fuel used, are tightly controlled and this results in a fuel economy measurement that is repeatable to within 72%, typically. While this measured fuel economy may differ signicantly from the fuel economy for the same vehicle in any specic use, the measured value provides a comparative benchmark for vehicle fuel economy that is useful from a vehicle buyers perspective and from an engineering perspective. The sales-weighted average test fuel economy of all new vehicles sold in the United States is about 28 mpg for cars and 21 mpg for light trucks. Much of this difference between car and light-truck fuel economy is attributable to the fact that light trucks are larger and heavier than cars, but some of the difference is also attributable to the fact that cars utilize higher levels of efciency enhancing technology. Fuel economy levels in Australia and Canada are similar to the U.S. levels, but cars in the EU have about 25% higher fuel economy, on average. The higher fuel economy in the EU is partly due to the smaller size and weight of cars sold and partly due to the higher penetration of diesel engines, which are more efcient than spark ignition engines.

Studies conducted by technical agencies have concluded that vehicle fuel economy can be increased substantially from average values without any reduction of attributes such as interior space of cargo carrying ability. The sources of energy loss and the technology available to reduce these losses are described later. A simple model of energy consumption in conventional automobiles provides insight into the sources and nature of energy losses. In brief, the engine converts fuel energy to shaft work. This shaft work is used to overcome the tractive energy required by the vehicle to move forward, as well as to overcome driveline losses and supply accessory drive energy requirements. The tractive energy can be separated into the energy required to overcome aerodynamic drag force, rolling resistance and inertia force. It is useful to consider energy consumption on the U.S. city and highway test cycles, which are reference cycles for comparing fuel economy. Denoting the average engine brake specic fuel consumption over the test cycle as bsfc, we have fuel consumption, FC, given by FC bsfc ER EA EK bsfc EAC Gi ti tb Zd

where Zd is the drive train efciency, ER is the energy to overcome rolling resistance, EA is the energy to overcome aerodynamic drag, EK is the energy to overcome inertia force, EAC is the accessory energy consumption, Gi is idle fuel consumption per unit time, and ti, tb are the time spent at idle and braking. The rst term in the above equation represents the fuel consumed to overcome tractive forces. Since the Federal Test Procedure (FTP) species the city and highway test cycle in terms of speed versus time, ER, EA, and Ek can be readily calculated as function of the vehicle weight, the tire rolling resistance, body aerodynamic drag coefcient, and vehicle frontal area. Weight reduction reduces both inertia force and rolling resistance. It should be noted that not all of the inertia force is lost to the brakes, as a vehicle will slow down without the use of brakes, at zero input power due to aerodynamic drag and rolling resistance. Braking energy loss is approximately 35% in the city cycle and 7% on highway cycle. The fuel energy is used not only to supply tractive energy requirements but also to overcome transmission losses, accounting for the transmission efciency that is in the rst term. The second term in the equation is for the fuel consumed to run the accessories. Accessory power requirements are required to run the radiator cooling

Internal Combustion Engine Vehicles

499

fan, alternator, water pump, oil pump, and power steering pump. Air conditioners also absorb power but are not reected in ofcial fuel economy estimates since they are not turned on during the FTP. Idle and braking fuel consumption are largely a function of engine size and idle RPM, while transmission losses are function of transmission type (manual or automatic) and design. The engine produces no power during idle and braking but consumes fuel, so that factor is accounted for by the third term. Table I shows the energy consumption (as a percentage) by all of these factors for a typical U.S. midsize car of mid-1990s vintage, with a 3-liter displacement s.i. engine, four-speed automatic transmission with lock-up, and power steering. The values in Table I can be utilized to derive sensitivity coefcients for the reduction of various loads. For example, reducing the weight by 10% will reduce both rolling resistance and inertia weight forces, so that tractive energy is reduced by (30.35 40.22) 0.1 or 7.06% on the composite cycle. Fuel consumption will be reduced by 7.06% 0.6544, which is the fraction of fuel used by tractive energy, or 4.6%. This matches the common wisdom that reducing weight by 10% reduces fuel consumption by 4 to 5%. However, if the engine is also downsized by 10% to account for the weight loss, fuel consumption will be reduced by 5.8% since idle and braking fuel consumption will be reduced in proportion to engine size. In addition, there will be some reduction (0.5%) in transmission and drivetrain loss. Fuel economy can be improved by two primary methods: (1) by reducing the power required to propel
TABLE I

the vehicle and (2) by increasing the engine efciency. To estimate the effects of different technology improvements that affect engine power required or the efciency of the engine, it is useful to keep certain vehicle attributes constant. Vehicle attributes of interest to consumers are passenger room, cargo space or payload capability, acceleration performance, and vehicle comfort/convenience features. The impact of technology on fuel economy is typically measured while keeping these attributes constant. Reducing the power required to propel the vehicle reduces engine load and can be accomplished by reduction of weight, aerodynamic drag, rolling resistance, or accessory loads. Engine efciency increases can be accomplished not only by engine technologies but also by improved drivetrain technologies that improve the match between engine operating point and vehicle power requirements. Spark ignition engines convert only about 20 to 25% of fuel energy to useful work during typical driving so that a doubling of engine efciency is theoretically possible without changing the basic Otto cycle.

3. IMPROVING EFFICIENCY BY REDUCING TRACTIVE ENERGY REQUIRED

Since vehicle weight is one of the most important variables determining fuel economy, weight reduction is an important method of improving fuel economy. The vehicles weight is distributed between the body structure, the drivetrain, the vehicles

Energy Consumption as a Percentage of Total Energy Requirements for a Typical Midsize Cara City Percentage of total tractive energy Rolling resistance Aerodynamic drag Inertia (Weight) force Total Percentage of total fuel consumed Tractive energy Accessory energy Idle Braking consumption Transmission Driveline loss 57.5 10.0 15.0 17.5 80.0 6.5 2.0 11.5 65.44 8.76 10.41 15.39 27.7 18.0 54.3 100 35.2 50.4 14.4 100 30.35 29.43 40.22 100 Highway Compositeb

a Midsize car of inertia weight 1588 kg, CD 0.33, A 2.1 m2, CR 0.011, 3L OHV V-6, power steering, four-speed automatic transmission with lock-up, air conditioning. b Highway fuel economy is 1.5 times city fuel economy, and composite gures are based on the U.S. EPA 55% city/45% highway fuel consumption weighting.

500

Internal Combustion Engine Vehicles

interior, and vehicle suspension/tires. The rst two component groups account for over 75% of a vehicles weight. Weight can be reduced in all four component groups by improved structural design as well as by the use of alternative materials. Improved structural design and packaging has been made possible through advanced computer simulations of structural strength, so that material use and shape can be optimized for the loads encountered. Most modern cars feature unibody designs where the body panels carry the structural loads, but several older models as well as many light trucks continue to use a separate chassis to carry structural loads. Heavy trucks, however, almost always utilize a separate chassis on which body components are mounted. A new architecture called space frame designs have emerged where structural loads are carried on skeletal frame from which body panels are hung. Improved packaging by optimization of component placement, body layout, and drivetrain location can also yield weight benets. The placement of the engine transversely between the front wheels and driving the front wheels provides signicant packaging benets over front engine, rear-wheel-drive packages for light-duty vehicles. The use of alternative materials such as ultra-highstrength steel, aluminum, and plastic composites is another way to reduce weight. Because of its low cost, steel, and cast iron continue to be the material of choice for body structures. Aluminum is already widely used for engine blocks and cylinder heads, and it is also used in critical suspension components. Some luxury cars now feature all-aluminum bodies, which weigh 30 to 35% less than their steel counterparts. Plastic composites are also widely used in body closures such as fenders, hood, and decklid with weight savings of 20 to 25% relative to steel parts. Such composites also see wide usage for lightweight interiors in the vehicle dashboard, seats, and door panels. Specially constructed prototypes maximizing the use of lightweight alternative materials have shown that weight reduction of 25 to 30% (relative to a conventional average steel vehicle) is possible, although with higher cost and with manufacturability constraints. The use of alternative materials in heavy trucks may not reduce loaded weight but will permit a larger payload to be carried. Aerodynamic drag can be reduced by styling the vehicles exterior shape and guiding the vehicles interior airow. At the speeds experienced by a typical vehicle, low drag shapes are a result of careful attention to airow at the front of the vehicle, rear wheel wells and outside mirrors, and at the end of the

roof. A measure of the drag is the aerodynamic drag coefcient, CD, which is dened as Drag force CD 1 2 ; 2rv A where r is the density of air, v is the velocity of airow, and A is the vehicle frontal area. In the early 1980s, cars had drag coefcients of 0.45 to 0.5. By 2000, the most aerodynamic cars had drag coefcient of 0.25 to 0.28. Trucks typically have higher drag coefcients because of their boxy shape and increased ground clearance, relative to cars. Prototype cars with drag coefcients as low as 0.15 have been built, but such designs typically involve reduction of vehicle attributes such as reduced rear passenger headroom, reduced rear visibility, or reduced cargo space. Nevertheless, drag reduction still offers opportunities to reduce fuel consumption. The tires rolling resistance is the third major contributor to overall load. The tire rolling resistance coefcient (CR) is a measure of tire energy loss, and is dened as T ; CR LR where T is the torque required at any speed, R the tire radius, and L the vertical load on the tire. Typically, most modern tires have a CR in the range of 0.009 to 0.012. The tire rolling resistance results from a combination of tire-to-road friction and hysteresis. As the tire deforms, heat is dissipated in the tires sidewall and tread due to the visco-elastic nature of rubber. In comparison, a steel wheel riding on steel rails has about one-tenth the rolling resistance of a rubber tire. Tire rolling resistance can be reduced by improved design of the tire tread, shoulders, and belts. In addition, the tire material formulation can signicantly reduce hysteresis loss. The use of silica compounds mixed with rubber has been found to reduce rolling resistance, without affecting other desirable properties such as braking and wet traction. Design improvements and changes in belt material are also capable of reducing CR with limited or no reduction of desirable attributes. It appears possible to reduce CR by 15 to 25% in the short term and by up to 40% over the long term (B25 years).

4. IMPROVEMENT OF ENGINE EFFICIENCY

Engine efciency on the driving cycle is the most signicant determinant of vehicle fuel economy for a

Internal Combustion Engine Vehicles

501

vehicle of a specic weight. Heat engine efciency can be stated in several ways. One intuitively appealing method is to express the useful energy produced by an engine as a percentage of the total heat energy that is theoretically liberated by combusting the fuel. This is sometimes referred to as the rst law efciency, implying that its basis is the rst law of thermodynamics, the law of conservation of energy. Another potential, but less widely used, measure is based on the second law of thermodynamics, which governs how much of that heat can be converted to work. Given a maximum combustion temperature (usually limited by engine material considerations and by emission considerations), the second law postulates a maximum efciency based on an idealized heat engine cycle called the Carnot cycle. The ratio of the rst law efciency to the Carnot cycle efciency can be utilized as a measure of how efciently a particular engine is operating with reference to the theoretical maximum based on the second law of thermodynamics. However, the most common measure of efciency used by automotive engineers is termed brake specic fuel consumption (bsfc), which is the amount of fuel consumed per unit time per unit of power. In the United States, the bsfc of engines is usually stated in pounds of fuel per brake horsepower hour, whereas the more common metric system measurement unit is in grams per kilowatt-hour (g/kwh). The term brake here refers to a common method historically used to measure engine shaft power output. Of course, all three measures of efciency are related to each other. The efciency of Otto and diesel cycle engines is not constant but depends on the operating point of the engine as specied by its torque output and shaft speed (revolutions per minute or RPM). Engine design considerations, frictional losses, and heat losses result in a single operating point where efciency is highest. This maximum efciency for an s.i. engine usually occurs at relatively high torque and at low to mid-RPM within the operating RPM range of the engine. At idle, the efciency is zero since the engine is consuming fuel but not producing any useful work. When considering efciency in a vehicle, the maximum efciency need not, by itself, be an indicator of the average efciency under normal driving conditions, since engine speed and torque vary widely under normal driving. The maximum efciency of an engine is of interest to automotive engineers, but a more practical measure of efciency is its average efciency during normal driving or during the ofcial city and highway fuel economy test.

4.1 Theoretical Maximum Engine Efciency

The characteristic features common to all piston internal combustion engines are as follows: 1. Intake and compression of the air or air-fuel mixture 2. Raising the temperature (and hence, the pressure) of the compressed air by combustion of fuel 3. The extraction of work from the high-pressure products of combustion by expansion 4. Exhaust of the products of combustion Combustion of the homogenous air-fuel mixture in a spark ignition engine takes place very quickly relative to piston motion and is represented in idealized calculations as an event occurring at constant volume. According to classical thermodynamic theory, the thermal efciency, Z, of an idealized Otto cycle, starting with intake air-fuel mixture drawn in at atmospheric pressure, is given by Z 1 1=rn1 ; 1

where r is the compression (and expansion) ratio and n is the ratio of specic heat at constant pressure to that at constant volume for the mixture. The equation shows that efciency increases with increasing compression ratio. Using an n value of 1.4 for air, the equation predicts an efciency of 58.47% at a compression ratio of 9:1. A value of n 1.26 is more correct for products of combustion of a stoichiometric mixture of air and gasoline. A stoichiometric mixture corresponds to an air-fuel ratio of 14.7:1, and this air-fuel ratio is typical for most spark ignition engines sold in the United States. At this air-fuel ratio, calculated efciency is about 43.5%. Actual engines yield still lower efciencies even in the absence of mechanical friction, due to heat transfer to the wall of the cylinder and the inaccuracy associated with assuming combustion to be instantaneous. Figure 1 shows the pressure-volume cycle of a typical spark ignition engine and its departure from the ideal relationship. Compression ratios are limited by the octane number of gasoline, which is a measure of its resistance to preignition or knock. At high compression ratios, the heat of compression of the air-fuel mixture becomes high enough to induce spontaneous combustion of small pockets of the mixture, usually those in contact with the hottest parts of the combustion chamber. These spontaneous combustion events are like small explosions that can damage the engine and

502

Internal Combustion Engine Vehicles

Pressure, psia 1000

900 800 700 600 500 400 300

Pressure, psia 1000

900 800 700 600 500 400

200

300 200

n = 1.28 n = 1.3

10090
80 70 60 50 40 30 20

n = 1.28 10090 n = 1.25

80 70 60 50 40 30 20

10 1

5 6 7 89

20

30

40 50 60 708090

10 Volume, cubic inches

100

10 1

5 6 7 89

20

30

40 50 60 708090

FIGURE 1 Pressure-volume diagram for a gasoline engine.

Compression ratio 8.711.

10 Volume, cubic inches

100

FIGURE 2 Pressure-volume diagram for a diesel engine. Compression ratio 1713.

can reduce efciency depending on when they occur during the cycle. Higher octane number gasoline prevents these events, but also costs more and requires greater energy expenditure for manufacture at the renery. The octane number is measured using two different procedures resulting in two different ratings for a given fuel, called motor octane and research octane number. Octane numbers displayed at the pump are an average of research and motor octane numbers, and most engines sold in the United States require regular gasoline with a pump octane number of 87. Typical compression rates for engines running on regular gasoline are in the range of 9:1 to 10:1. The diesel, or compression ignition, engine differs from the spark ignition engine in that only air, rather than the air-fuel mixture, is compressed. The diesel fuel is sprayed into the combustion chamber at the end of compression in a ne mist of droplets and the diesel fuel ignites spontaneously upon contact with the compressed air due to the heat of compression. The sequence of processes (i.e., intake, compression, combustion, expansion, and exhaust) is similar to that of an Otto cycle engine. However, the combustion process occurs over a relatively long period and is represented in idealized calculations as an event occurring at constant pressure (i.e., combustion occurs as the piston moves downward to increase volume and decrease pressure at a rate offsetting the pressure rise due to heat release). Figure 2 shows the pressurevolume cycles for a typical diesel engine and its rela-

tionship to the ideal diesel cycle. If the ratio of volume at the end of the combustion period to the volume at the beginning of the period is rc, or the cutoff-ratio, the thermodynamic efciency of the idealized constant-pressure combustion cycle is given by  n  1 rc 1 Z n 1 : 2 r nrc 1 It can be seen that for rc 1, the combustion occurs at constant volume and the efciency of the diesel and Otto cycle are equivalent. The term rc also measures the interval during which fuel is injected, and it increases as the power output is increased. The efciency equation shows that as rc is increased, efciency falls so that the idealized diesel cycle is less efcient at high loads. The combustion process also is responsible for a major difference between diesel and Otto cycle engines. In an Otto cycle engine, intake is air throttled to control power while maintaining a near constant air-fuel ratio; in a diesel engine, power control is achieved by varying the amount of fuel injected while keeping the air mass inducted per cycle at near constant levels. In most operating modes, combustion occurs with considerable excess air in a c.i. engine, while combustion occurs at or near stoichiometric air-fuel ratios in a modern s.i. engine.

Internal Combustion Engine Vehicles

503

At the same compression ratio, the Otto cycle has the higher efciency. However, diesel cycle engines normally operate at much higher compression ratios, since there are no octane limitations associated with this cycle. In fact, spontaneous combustion of the fuel is required in such engines, and the ease of spontaneous combustion is measured by a fuel property called cetane number. Most c.i. engines require diesel fuels with a cetane number over 40. In practice, there are two kinds of c.i. engines, the direct injection type (DI) and the indirect injection type (IDI). The DI type utilizes a system where fuel is sprayed directly into the combustion chamber. The fuel spray is premixed and partially combusted with air in a prechamber in the IDI engine, before the complete burning of the fuel in the main combustion chamber occurs. DI engines generally operate at compression ratios of 15 to 20:1, while IDI engines operate at 18 to 23: 1. The theoretical efciency of a c.i. engine with a compression ratio of 20:1, operating at a cutoff ratio of 2, is about 54% (for combustion with excess air, n is approximately 1.3). In practice, these high efciencies are not attained, for reasons similar to those outlined for s.i. engines.

4.2 Actual versus Theoretical Efciency

Four major factors affect the efciency of s.i. and c.i. engines. First, the ideal cycle cannot be replicated due to thermodynamic and kinetic limitations of the combustion process, and the heat transfer that occurs from the cylinder walls and combustion chamber. Second, mechanical friction associated with the motion of the piston, crankshaft, and valves consume a signicant fraction of total power. Since friction is a stronger function of engine speed rather than torque, efciency is degraded considerably at light load and high RPM conditions. Third, aerodynamic frictional losses associated with airow through the air cleaner, intake manifold and valves, exhaust manifold, silencer, and catalyst are signicant, especially at high airow rates through the engine. Fourth, pumping losses associated with throttling the airow to achieve part-load conditions in spark ignition engines are very high at light loads. Note that c.i. engines do not usually have throttling loss, and their part load efciencies are superior to those of s.i. engines. Efciency varies with both speed and load for both engine types. Hence, production spark ignition or compression ignition engines do not attain the theoretical values of efciency, even at their most efcient operating point. In general, for both types of engines, the

maximum efciency point occurs at an RPM that is intermediate to idle and maximum RPM and at a level that is 60 to 75% of maximum torque. On-road average efciencies of engines used in cars and light trucks are much lower than peak efciency, since the engines generally operate at very light loads during city driving and steady-state cruise on the highway. High power is utilized only during strong accelerations, at very high speeds, or when climbing steep gradients. The high load conditions are relatively infrequent, and the engine operates at light loads much of the time during normal driving. During normal driving, the heat of fuel combustion is lost to a variety of sources and only a small fraction is converted to useful output, resulting in the low values for on-road efciency. Figure 3 provides an example of the heat balance for a typical modern small car with a spark ignition engine under a low speed (25 mph or 40 mph) and a high-speed (62 mph or 100 mph) condition. At very low driving speeds typical of city driving, most of the heat energy is lost to the engine coolant. Losses associated with other waste heat include radiant and convection losses from the hot engine block and heat losses to the engine oil. A similar heat loss diagram for a diesel c.i. would indicate lower heat loss to the exhaust and

Other waste heat 19.8% QI Le

Engine output 20.8%

Qex Waste heat of exhaust gas 24.8% 40 km/h Other waste heat 26.3% QI Le Engine output 28.3% Qw Waste heat of engine cooling 34.7%

Qex Waste heat of exhaust gas 18.2%

Qw

Waste heat of engine cooling 27.3%

100 km/h

FIGURE 3 Heat balance of a passenger car equipped with

1500cc engine 17.

504
500 Specific fuel consumption (g/kW-h)

Internal Combustion Engine Vehicles

Friction mean effective pressure Brake Constant speed and air-fuel ratio 3.0

bar

400
Fr
s

O il
Va lve tra

os

se

pu m p
in

300

P um

pin g

lo s s es

G Wa ene r t e r a t or pu mp

i ct
io

1.5 200 Indicated

Pi st on rin gs

Engine friction distribution

20

40 60 Percentage of brake load

80

100

Pi

st

on

FIGURE 4 Specic fuel consumption versus engine load 18.

ne Con

cting

rod

coolant and an increased fraction of heat converted to work, especially at the low speed condition. During stop-and-go driving conditions typical of city driving, efciencies are even lower than those indicated in Fig. 3 because of the time spent at idle where efciency is zero. Under the prescribed U.S. city cycle conditions, typical modern spark ignition engines have an efciency of about 18%, modern IDI c.i. engines have an efciency of about 21%, and modern DI diesel have an efciency of about 23%. Another method of examining the energy losses is by allocating the power losses starting from the power developed within the cylinder. The useful work corresponds to the area that falls between the compression and expansion curve depicted in Figs 1 and 2. The pumping work that is subtracted from this useful work, referred to as indicated work, is a function of how widely the throttle is open and, to a lesser extent, the speed of the engine. Figure 4 shows the dependence of specic fuel consumption (or fuel consumption per unit of work) with load, at constant (low) engine RPM. Pumping work represents only 5% of indicated work at full load, low RPM conditions, but increases to over 50% at light loads of less than two-tenths of maximum power. Mechanical friction and accessory drive power, on the other hand, increase nonlinearly with engine speed but do not change much with the throttle setting. Figure 5 shows the contribution of the various engine components as well as the alternator, water pump and oil pump to total friction, expressed in terms of mean effective pressure, as a function of RPM. The brake mean effective pressure is

Crankshaft bearing 0.0 0 700 (Idle) 3000 6000 Speed RPM

FIGURE 5 Friction distribution as a function of engine speed

19.

a measure of specic torque, or torque per unit of engine displacement; typical engine brake mean effective pressure (bmep) of spark ignition engines that are not supercharged range from 8.5 to 10 bar. Hence, friction accounts for about 25% of total indicated power at high RPM (B6000) but only for about 10% of indicated power at low RPM (B2000) in spark ignition engines. Friction in a c.i. engine is higher because of the need to maintain an effective pressure seal at high compression ratios, and the friction mean effective pressure is 30 to 40% higher than that for a dimensionally similar s.i. engine at the same RPM. Since the brake mean effective pressure of a diesel is also lower than that of a gasoline engine, friction accounts for 15 to 16% of indicated maximum power even at 2000 RPM. Typical bmep values for a naturally aspirated c.i. engine range from 6.5 to 7.5 bar.

5. INCREASING THE EFFICIENCY OF SPARK IGNITION ENGINES

5.1 Design Parameters
Engine valvetrain design is a widely used method to classify spark ignition engines. The rst spark

Internal Combustion Engine Vehicles

505

ignition engines were of the side-valve type, but such engines have not been used in automobiles for several decades, although some engines used in off-highway applications, such as lawn mowers or forklifts, continue to use this design. The overhead valve (OHV) design supplanted the side-valve engine by the early 1950s, and continues to be used in many U.S. engines in much improved form. The overhead cam engine (OHC) is the dominant design used in the rest of the developed world. The placement of the camshaft in the cylinder heads allows the use of simple, lighter valvetrain, and valves can be opened and closed more quickly as a result of the reduced inertia. This permits better ow of intake and exhaust gases, especially at high RPM, with the result that an OHC design typically can produce greater power at high RPM than an OHV design of the same displacement. A more sophisticated version of the OHC engine is the double overhead cam (DOHC) engine where two separate camshafts are used to activate the intake and exhaust valves, respectively. The DOHC design permits a very light valvetrain as the camshaft can actuate the valves directly without any intervening mechanical linkages. The DOHC design also allows some layout simplication, especially in engines that feature two intake valves and two exhaust valves (4-valve). The 4-valve engine has become popular since the mid-1980s and Japanese manufacturers, in particular, have embraced the DOHC 4-valve design. The DOHC design permits higher specic output than an OHC design, with the 4-valve DOHC design achieving the highest specic output, in excess of 70 BHP/liter of displacement.

5.2 Thermodynamic Efciency

Increases in thermodynamic efciency within the limitations of the Otto cycle are obviously possible by increasing the compression ratio. However, compression ratio is also fuel octane limited, and increases in compression ratio depend on how the characteristics of the combustion chamber and the timing of the spark can be tailored to prevent knock while maximizing efciency. Spark timing is associated with the delay in initiating and propagating combustion of the air-fuel mixture. To complete combustion before the piston starts its expansion stroke, the spark must be initiated a few crank angle degrees (advance) before the piston reaches top dead center. For a particular combustion chamber, compression ratio, and air-fuel mixture, there is an optimum level of spark advance for maximizing combustion chamber pressure and,

hence, fuel efciency. This level of spark advance is called MBT for maximum for best torque. However, MBT spark advance can result in knock if fuel octane is insufcient to resist preignition at the high pressures achieved with this timing. Hence, there is an interplay between spark timing and compression ratio in determining the onset of knock. Retarding timing from MBT reduces the tendency to knock but decreases fuel efciency. Emissions of hydrocarbons and oxides of nitrogen (NOx) are also dependent on spark timing and compression ratio, so that emission constrained engines require careful analysis of the knock, fuel efciency, and emission trade-offs before the appropriate value of compression ratio and spark advance can be selected. Electronic control of spark timing has made it possible to set spark timing closer to MBT relative to engines with mechanical controls. Due to production variability and inherent timing errors in a mechanical ignition timing system, the average value of timing in mechanically controlled engines had to be retarded signicantly from the MBT timing. This protects the fraction of engines with higher than average advance due to production variability from knock. The use of electronic controls coupled with magnetic or optical sensors of crankshaft position has reduced the variability of timing between production engines and also allowed better control during transient engine operation. Engines have been equipped with knock sensors, which are essentially vibration sensors tuned to the frequency of knock. These sensors allow for advancing ignition timing to the point where trace knock occurs, so that timing is optimal for each engine produced regardless of production variability. High-swirl, fast-burn combustion chambers have been developed to reduce the time taken for the airfuel mixture to be fully combusted. The shorter the burn time, the more closely the cycle approximates the theoretical Otto cycle with constant volume combustion and the greater the thermodynamic efciency. Reduction in burn time can be achieved by having a turbulent vortex within the combustion chamber that promotes ame propagation and mixing. The circular motion of the air-fuel mixture is known as swirl, and turbulence is also enhanced by shaping the piston so that gases near the cylinder wall are pushed rapidly towards the center in a motion known as squish. Improvements in ow visualization and computational uid dynamics have allowed the optimization of intake valve, inlet port, and combustion chamber geometry to achieve desired ow characteristics. Typically, these designs

506

Internal Combustion Engine Vehicles

have resulted in a 2 to 3% improvement in thermodynamic efciency and fuel economy. The high-swirl chambers also allow higher compression ratios and reduced spark advance at the same fuel octane number. The use of these types of combustion chambers has allowed compression ratio from about 8:1 in the early 1980s to 10:1 in the early 2000s, and further improvements are likely. In newer engines of the 4-valve DOHC type, the sparkplug is placed at the center of the combustion chamber, and the chamber can be made very compact by having a nearly hemispherical shape. Engines incorporating these designs have compression ratios of 10:1 while still allowing the use of regular 87 octane gasoline. Increases beyond 10:1 are expected to have diminishing benets in efciency and fuel economy and compression ratios beyond 12:1 are not likely to be benecial unless fuel octane is raised simultaneously.

5.3 Reduction in Mechanical Friction

Mechanical friction losses are being reduced by converting sliding metal contacts to rolling contacts, reducing the weight of moving parts, reducing production tolerances to improve the t between pistons and bore, and improving the lubrication between sliding or rolling parts. Friction reduction has focused on the valvetrain, pistons, rings, crankshaft, crankpin bearings, and the oil pump. Valvetrain friction accounts for a larger fraction of total friction losses at low engine RPM than at high RPM. The sliding contract between the cam that activates the valve mechanism through a pushrod in an OHV design, or a rocker arm in an OHV design, can be substituted with a rolling contact by means of a roller cam follower. Roller cam followers have been found to reduce fuel consumption by 2 to 4% during city driving and 1 to 2% in highway driving. The use of lightweight valves made of ceramics or titanium is another possibility for the future. The lightweight valves reduce valve train inertia and also permit the use of lighter springs with lower tension. Titanium alloys are also being considered for valve springs which operate under heavy loads. There alloys have only half the shear modulus of steel and fewer coils are needed to obtain the same spring constant. A secondary benet associated with lighter valves and springs is that the erratic valve motion at high RPM is reduced, allowing increased engine RPM range and power output. The pistons and rings contribute to approximately half of total friction. The primary function of the rings is to minimize leakage of the air-fuel mixture

from the combustion chamber to the crankcase, and oil leakage from the crankcase to the combustion chamber. The ring pack for most engines is composed of two compression rings and an oil ring. The rings have been shown to operate hydrodynamically over the cycle, but metal-to-metal contact occurs often at the top and bottom of the stroke. The outward radial force of the rings are as a result of installed ring tension and contribute to effective sealing as well as friction. Various low-tension ring designs were introduced in the 1980s, especially since the need to conform to axial diameter variations or bore distortions have been reduced by improved cylinder manufacturing techniques. Reduced tension rings have yielded friction reduction in the range of 5 to 10%, with fuel economy improvements of 1 to 2%. Elimination of one of the two compression rings has also been tried on some engines, and two-ring pistons may be the low friction concept for the future. Pistons have also been redesigned to decrease friction. Prior to the 1980s, piston had large skirts to absorb side forces associated with side-to-side piston motion due to engine manufacturing inaccuracies. Pistons with reduced skirts diminish friction by having lower surface area in contact with the cylinder wall, but this effect is quite small. A larger effect is obtained for the mass reduction of a piston with smaller skirts and piston skirt size has seen continuous reduction since the 1980s. Reducing the reciprocating mass reduces the piston-to-bore loading. Secondary benets include reduced engine weight and reduced vibration. Use of advanced materials also result in piston weight reduction. Lightweight pistons use hypereutectic aluminum alloys, while future pistons could use composite materials such as ber-reinforced plastics. Advanced materials can also reduce the weight of the connecting rod, which also contributes to the side force on a piston. The crankshaft bearings include the main bearings that support the crankshaft and the crankpin bearings and are of the journal bearing type. These bearings contribute to about 25% of total friction, while supporting the stresses transferred from the piston. The bearings run on a lm of oil and detailed studies of lubrication requirements has led to optimization of bearing width and clearances to minimize engine friction. Studies on the use of roller bearings rather than journal bearings in this application has shown further reduction in friction is possible. Crankshaft roller bearings are used only in some two-stroke engines such as outboard motors for boat propulsion, but their durability in automotive applications has not been established.

Internal Combustion Engine Vehicles

507

Coatings of the piston and ring surfaces with materials to reduce wear also contribute to friction reduction. The top ring, for example, is normally coated with molybdenum, and new proprietary coating materials with lower friction are being introduced. Piston coatings of advanced high temperature plastics or resin are used widely and are claimed to reduce friction by 5% and fuel consumption by 1%. The oil pumps generally used in most engines are of the gear pump type. Optimization of oil ow rates and reduction of the tolerances for the axial georotor clearance has led to improved efciency, which translates to reduced drive power. Friction can be reduced by 2 to 3% with improved oil pump designs for a gain in fuel economy of about half of a percent. Improvements to lubricants used in the engine also contribute to reduced friction and improved fuel economy. There is a relationship between oil viscosity, oil volatility, and engine oil consumption. Reduced viscosity oils traditionally resulted in increased oil consumption, but the development of viscosity index (VI) improvers had made it possible to tailor the viscosity with temperatures to formulate multigrade oils such as 10W-40 (these numbers refer to the range of viscosity covered by a multigrade oil). These multigrade oils act like low-viscosity oils during cold starts of the engine, reducing fuel consumption, but retain the lubricating properties of higher viscosity oils after the engine warms up to normal operating temperature. The development of 5W-20 oils and 5W-40 oils can contribute to a fuel economy improvement by further viscosity reduction. Friction modiers containing molybdenum compounds have also reduced friction without affecting wear or oil consumption. Future synthetic oils combining reduced viscosity and friction modiers can offer good wear protection, low oil consumption, and extended drain capability along with small improvements to fuel economy in the range of 1 to 3% over current oils.

5.4 Reduction in Pumping Loss

Reductions in ow pressure loss can be achieved by reducing the pressure drop that occurs in the ow of air (air-fuel mixture) into the cylinder and the combusted mixture through the exhaust system. However, the largest part of pumping loss during normal driving is due to throttling, and strategies to reduce throttling loss have included variable valve timing and lean-burn systems. The pressure losses associated with the intake system and exhaust system have been typically

dened in terms of volumetric efciency, which is a ratio of the actual airow through an engine to the airow associated with lling the cylinder completely. The volumetric efciency can be improved by making the intake airow path as free of ow restrictions as possible through the air lters, intake manifolds, and valve ports. The shaping of valve ports to increase swirl in the combustion chamber can lead to reduced volumetric efciency, leading to a trade-off between combustion and volumetric efciency. More important, the intake and exhaust processes are transient in nature as they occur only over approximately half a revolution of the crankshaft. The momentum effects of these ow oscillations can be exploited by keeping the valves open for durations greater than half a crankshaft revolution. During the intake stroke, the intake valve can be kept open beyond the end of the intake stroke, since the momentum of the intake ow results in a dynamic pressure that sustains the intake ow even when the piston begins the compression stroke. A similar effect is observed in the exhaust process, and the exhaust valve can be held open during the initial part of the intake stroke. These ow momentum effects depend on the velocity of the ow which is directly proportional to engine RPM. Increasing the valve opening duration helps volumetric efciency at high RPM but hurts it at low RPM. Valve timing and overlap are selected to optimize the trade-off between high and low RPM performance characteristics. Efciency improvements can be realized by changing the valve overlap period to provide less overlap at idle and low engine speeds and greater overlap at high RPM. In DOHC engines, where separate crankshafts actuate the intake and exhaust valves, the valve overlap period can be changed by rotating the camshafts relative to each other. Such mechanisms have been commercialized engines show low RPM torque improvements of 7 to 10% with no sacrice in maximum horsepower attained in the 5500 to 6000 RPM range. Variable valve overlap period is just one aspect of a more comprehensive variable valve timing system. The oscillatory intake and exhaust ows can allow volumetric efciency to be increased by exploiting resonance effects associated with pressure waves similar to those in organ pipes. The intake manifolds can be designed with pipe lengths that resonate, so that a high-pressure wave is generated at the intake valve as it is about to close, to cause a supercharging effect. Exhaust manifolds can be designed to resonate to achieve the opposite pressure effect to purge exhaust gases from this cylinder. For a given pipe

508

Internal Combustion Engine Vehicles

length, resonance occurs only at a certain specic frequency and its integer multiples so that, historically, tuned intake and exhaust manifolds could help performance only in certain narrow RPM ranges. The incorporation of a resonance tanks using the Helmholtz resonator principle in addition to tuned length intake pipes has led to improved intake manifold design that provide benets over broader RPM ranges. Variable resonance systems have been introduced, where the intake tube lengths are changed at different RPM by opening and closing switching valves to realize smooth and high torque across virtually the entire engine speed range. Typically, the volumetric efciency improvement is in the range of 4 to 5% over xed resonance systems. Another method to increase efciency is by increasing valve area. A 2-valve design is limited in valve size by the need to accommodate the valves and sparkplugs in the circle dened by the cylinder base. The active ow area is dened by the product of valve circumference and lift. Increasing the number of valves is an obvious way to increase total valve area and ow area, and the 4-valve system, which increases ow area by 25 to 30% over 2-valve layouts, has gained broad acceptance. The valves can be arranged around the cylinder bore and the sparkplug placed in the center of the bore to improve combustion. Analysis of additional valve layout designs that take into account the minimum required clearance between valve seats and the sparkplug location suggest that vevalve designs (3 intake, 2 exhaust) can provide an additional 20% increase in ow area, at the expense of increased valvetrain complexity. Additional valves do not provide further increases in ow area either due to noncentral plug locations or valve-to-valve interference. Under most normal driving conditions, the throttling loss is the single largest contributor to reduction in engine efciency. In s.i. engines, the air is throttled ahead of the intake manifold by means of a buttery valve that is connected to the accelerator pedal. The vehicles driver demands a power level by depressing or releasing the accelerator pedal, which in turn opens or closes the buttery valve. The presence of the buttery valve in the intake air stream creates a vacuum in the intake manifold at part throttle conditions, and the intake stroke draws in air at reduced pressure, resulting in pumping losses. These losses are proportional to the intake vacuum and disappear at wide open throttle. Measures to reduce throttling loss are varied. The horsepower demand by the driver can be satised by

any combination of torque and RPM since Power Torque RPM: The higher the torque, the lower the RPM to satisfy a given power demand. Higher torque implies less throttling, and the lower RPM also reduces friction loss so that the optimum theoretical fuel efciency at a given level of horsepower demand occurs at the highest torque level the engine is capable of. In practice, the highest level is never chosen because of the need to maintain a large reserve of torque for immediate acceleration and also because engine vibrations are a problem at low RPM, especially near or below engine speeds referred to as lugging RPM. Nevertheless, this simple concept can be exploited to the maximum by using a small displacement high specic output engine in combination with a multispeed transmission with ve or more forward gears. The larger number of gears allows selection of the highest torque/lowest RPM combination for fuel economy at any speed and load, while maintaining sufcient reserve torque for instantaneous changes in power demand. A specic torque increase of 10% can be utilized to provide a fuel economy benet of 3 to 3.5% if the engine is downsized by 8 to 10%. In light vehicles, the number of forward gears has been increasing from 3 to 5, and six-speed transmissions are likely to be standard in the future. The continuously variable transmission is other development that allow continuous change of gear ratios over a specic range. Lean-burn is another method to reduce pumping loss. Rather than throttling the air, the fuel ow is reduced so that the air-fuel ratio increases, or becomes leaner. (In this context, the c.i. engine is a lean-burn engine.) Most s.i. engines, however, do not run well at air: fuel ratios leaner than 18:1, as the combustion quality deteriorates under lean con;ditions. Engines constructed with high swirl and turbulence in the intake charge can run well at air: fuel ratios up to 21:1. In a vehicle, lean-burn engines are calibrated lean only at light loads to reduce throttling loss, but run at stoichiometric or rich air: fuel ratios at high loads to maximize power. The excess air combustion at light loads has the added advantage of having a favorable effect on the polytropic coefcient, n, in the efciency equation. Modern lean-burn engines do not eliminate throttling loss, but the reduction is sufcient to improve vehicle fuel economy by 8 to 10%. The disadvantage of lean burn is that such engines cannot yet use catalytic controls to reduce emissions of oxide of nitrogen (NOx), and the in-cylinder NOx emission

Internal Combustion Engine Vehicles

509

control from running lean is sometimes insufcient to meet stringent NOx emissions standards. However, there are developments in lean NOx catalysts that could allow lean-burn engines to meet the most stringent NOx standards proposed in the future. Another type of lean-burn s.i. engine is the stratied charge engine. Research is focused on direct-injection stratied charge (DISC) engines where the fuel is sprayed into the combustion chamber, rather than into or ahead of the intake valve. Typically, this enables the air: fuel ratio to vary axially or radially in the cylinder, with the richest air: fuel ratios present near the sparkplug or at the top of the cylinder. Stratication requires very careful design of the combustion chamber shape and intake swirl, as well as of the fuel injection system. Advanced direct injection systems have been able to maintain stable combustion at total air: fuel ratios as high as 40:1. Such engines have been commercialized in 2000 in Europe and Japan. Variable valve timing is another method to reduce throttling loss. By closing the intake valve early, the intake process occurs over a smaller fraction of the cycle, resulting in a lower vacuum in the intake manifold. It is possible to completely eliminate the buttery valve that throttles air and achieve all part load settings by varying the intake valve opening duration. However, at very light load, the intake valve is open for a very short duration, and this leads to weaker in-cylinder gas motion and reduced combustion stability. At high RPM, the throttling loss benets are not realized fully. Throttling occurs at the valve when the valve closing time increases relative to the intake stroke duration at high speeds, due to the valvetrain inertia. Hence, throttling losses can be decreased by 80% at light load, low RPM conditions, but by only 40 to 50% at high RPM, even with fully variable valve timing. Variable valve timing can also provide a number of other benets, such as reduced valve overlap at light loads/low speeds (discussed earlier) and maximized output over the entire range of engine RPM. Fully variable valve timing can result in engine output levels of up to 100 BHP/liter at high RPM with little or no effect on low-speed torque. In comparison to an engine with xed valve timing that offers equal performance, fuel efciency improvements of 7 to 10% are possible. The principal drawback has historically been the lack of a durable and low-cost mechanism to implement valve timing changes. A number of new systems have been introduced that are ingenious mechanisms with the required durability.

6. INCREASING THE EFFICIENCY OF COMPRESSION IGNITION (DIESEL) ENGINES

Compression ignition engines, commonly referred to as diesel engines, are in widespread use. Most c.i. engines in light-duty vehicle applications are of the indirect injection type (IDI), while most c.i. engines in heavy-duty vehicles are of the direct injection type. In comparison to s.i. engines, c.i. engines operate at much lower brake mean effective pressures of (typically) about 7 to 8 bar at full load. Maximum power output of a c.i. engine is limited by the rate of mixing between the injected fuel spray and hot air. At high fueling levels, inadequate mixing leads to high black smoke, and the maximum horsepower is usually smoke limited for most c.i. engines. Naturally aspirated diesel engines for light-duty vehicle use have specic power outputs of 25 to 35 BHP per liter, which is about half the specic output of a modern s.i. engine. However, fuel consumption is signicantly better, and c.i. engines are preferred over s.i. engines where fuel economy is important. Due to the combustion process, as well as the high internal friction of a c.i. engine, maximum speed is typically limited to less than 4500 RPM, which partially explains the lower specic output of c.i. engine. In light-duty vehicle use, an IDI engine can display between 20 to 40% better fuel economy depending on whether the comparison is based on engines of equal displacement or of equal power output in the same RPM range. The improvement is largely due to the superior part load efciency of the c.i. engine, as there is no throttling loss. At high vehicle speeds (4120 km/hr), the higher internal friction of the c.i. engine offsets the reduced throttling loss, and the fuel efciency difference between s.i. and c.i. engine narrows considerably. Most of the evolutionary improvements for compression ignition engines in friction and pumping loss reduction are conceptually similar to those described for s.i. engines, and this section focuses on the unique aspects of c.i. engine improvements.

6.1 Design Parameters

Note that c.i. engines have also adopted some of the same valvetrain designs as those found in s.i. engines. While most c.i. engines were of the OHV type, European c.i. engines for passenger car use are of the OHC type. The c.i. engine is not normally run at high RPM, so that the difference in specic output

510

Internal Combustion Engine Vehicles

between an OHV and an OHC design is small. The OHC design does permit a simpler and lighter cylinder block casting, which is benecial for overcoming some of the inherent weight liabilities. OHC designs also permits the camshaft to directly activate the fuel injector in unit injector designs, which are capable of high injection pressure and ne atomization of the fuel spray. Four-valve OHC or DOHC designs allow central placement of the fuel injector in the cylinder, which enhances uniform mixing of air and fuel.

6.2 Thermodynamic Efciency

The peak efciency of an IDI engine is comparable to or only slightly better than the peak efciency of an s.i. engine, based on average values for engines in production. The contrast between theoretical and actual efciency is notable, and part of the reason is that the prechamber in the IDI diesel is a source of energy loss. The design of the prechamber is optimized to promote swirl and mixing of the fuel spray with air, but the prechamber increases total combustion time. Its extra surface area also results in more heat transfer into the cylinder head. Direct injection (DI) systems avoid the heat and ow losses from the prechamber by injecting the fuel into the combustion chamber. The combustion process in DI diesels consists of two phases. The rst phase consists of an ignition delay period followed by spontaneous ignition of the fuel droplets. The second phase is characterized by diffusion burning of the droplets. The fuel injection system must be capable of injecting very little fuel during the rst phase and provide highly atomized fuel and promote intensive mixing during the second. Historically, the mixing process has been aided by creating high swirl in the combustion chamber to promote turbulence. However, high swirl and turbulence also lead to ow losses and heat losses, thus reducing efciency. The newest concept is the quiescent chamber where all of the mixing is achieved by injecting fuel at very high pressures to promote ne atomization and complete penetration of the air in the combustion chamber. New fuel injection systems using unit injectors can achieve pressures in excess of 1500 bar, twice as high as injection pressures utilized previously. Quiescent combustion chamber designs with high-pressure fuel injection systems have provided to be very fuel efcient and are coming into widespread use in heavy-duty truck engines. These systems have the added advantage of reducing particulate and smoke emissions.

DI engines have entered the light-duty vehicle market, but these engines still utilize swirl type combustion chambers. In combination with turbocharging (see Section 6), the new DI engines have attained peak efciencies of over 41%. Fuel economy improvements in the composite cycle relative to IDI engines are in the 12 to 15% range, and are up to 40% higher than naturally aspirated s.i. engines with similar torque characteristics. It is not clear if quiescent combustion chambers will be ever used in DI engines for cars, since the size of the chamber is quite small and fuel impingement on cylinder walls is a concern. Although the efciency equation shows that increasing compression ratio has a positive effect on efciency, practical limitations preclude any signicant efciency gain through this method. At high compression ratios, the size of the combustion chamber is reduced, and the regions of dead air trapped between the cylinder and piston edges and crevices became relatively large, leading to poor air utilization, reduced specic output, and, potentially, higher smoke. Moreover, the stresses on the engine increase with increasing compression ratio, making the engine heavy and bulky. The compression ratios are already somewhat higher than optimal to provide enough heat of compression so that a cold start at low ambient temperature is possible.

6.3 Friction and Pumping Loss

Most of the friction reducing technologies that can be adopted in s.i. engines are conceptually similar to those that can be adopted for diesels. There are limitations to the extent of reduction of ring tension and piston size due to the high compression ratio of c.i. engines, but roller cam followers, optimized crankshaft bearings, and multigrade lubricants have also been adopted for c.i. engine use. Since friction is a larger fraction of total loss, a 10% reduction in fraction in a c.i. engine can lead to a 3 to 4% improvement in fuel economy. Pumping losses are not as signicant a contributor to overall energy loss in a c.i. engine, but tuned intake manifolds and improved valve port shapes and valve designs have also improved volumetric efciency of modern c.i. engines. Four-valve designs, in widespread use in the heavy truck market, have appeared in passenger cars, but their benets are smaller in c.i. engine use due to the low maximum RPM relative to s.i. engines. Nevertheless, the 4-valve head with a centrally mounted injector is particularly useful in DI engines since it allows for

Internal Combustion Engine Vehicles

511

symmetry in the fuel spray with resultant good air utilization. Variable valve timing or any form of valve control holds little benet for c.i. engines due to the lack of throttling loss, and lack of high RPM performance. Valve timing can be varied to reduce the effective compression ratio, so that a very high ratio can be used for cold starts, but a lower, more optimal, ratio for fully warmed up operation.

mance vehicles. In c.i. engines, charge boosting is effective at all speeds and levels.

7.1 Turbochargers
Turbochargers in automotive applications are of the radial ow turbine type. The turbine extracts pressure energy from the exhaust stream and drives a compressor that increases the pressure of the intake air. A number of issues affect the performance of turbomachinery, some of which are a result of natural laws governing the interrelationship between pressure, airow, and turbocharger speed. Turbochargers do not function at light load because there is very little energy in the exhaust stream. At high load, the turbochargers ability to provide boost is a nonlinear function of exhaust ow. At low engine speed and high load, the turbocharger provides little boost, but boost increases rapidly beyond a certain ow rate that is dependent on the turbocharger size. The turbocharger also has a maximum ow rate, and the matching of a turbochargers ow characteristics to a piston engines ow requirements involves a number of trade-offs. If the turbocharger is sized to provide adequate charge boost at moderate engine speeds, high RPM boost is limited and there is a sacrice in maximum power. A larger turbocharger capable of maximizing power at high RPM sacrices the ability to provide boost at normal driving conditions. At very low RPM (for example, when accelerating from a stopped condition), no practical design provides boost immediately. Moreover, the addition of turbocharger requires the engine compression ratio to be decreased by 1.5 to 2 points (or 1 to 1.5 with an aftercooler) to prevent detonation. The net result is that turbocharged engines have lower brake specic fuel efciencies than engines of equal size, but can provide some efciency benet when compared to engines of equal mid range or top end power. During sudden acceleration, the turbocharger does not provide boost instantaneously due to its inertia, and turbocharged vehicles can have noticeably different acceleration characteristics than naturally aspirated vehicles. New variable geometry turbochargers have improved response and better boost characteristics over the operating range. Turbochargers are much better suited to c.i. engines since these engines are unthrottled and the combustion process is not knock limited. Airow at a given engine load/speed setting is always higher for a c.i. engine relative to an s.i. engine, and this provides a less restricted operating regime for the turbocharger. The lack of a knock limit also allows increased

7. INTAKE CHARGE BOOSTING

Most c.i. and s.i. engines for light vehicle use intake air at atmospheric pressure. One method to increase maximum power at wide open throttle is to increase the density of air supplied to the intake by precompression. This permits a smaller displacement engine to be substituted without loss of power and acceleration performance. The use of a smaller displacement engine reduces pumping loss at part load and friction loss. However, intake charge compression has its own drawbacks. Its effect is similar to raising the compression ratio in terms of peak cylinder pressure, but maximum cylinder pressure is limited in s.i. engines by the fuel octane. Charge boosted engines generally require premium gasolines with higher octane number if the charge boost levels are more than 0.2 to 0.3 bar over atmospheric in vehicles for street use. Racing cars use boost levels up to 1.5 bar in conjunction with a very high octane fuel such as methanol. This limitation is not present in a c.i. engine, and charge boosting is much more common in c.i. engine applications. Most c.i. engines in heavy-duty truck applications use charge boosting. Intake charge boosting is normally achieved by the use of turbochargers or superchargers. Turbochargers recover the wasted heat and pressure in the exhaust through a turbine, which in turn drives a compressor to boost intake pressure. Superchargers are generally driven by the engine itself and are theoretically less efcient than a turbocharger. Many engines that use either device also utilize an aftercooler that cools the compressed air as it exits from the supercharger or turbocharger before it enters the s.i. engine. The aftercooler increases engine specic power output by providing the engine with a denser intake charge, and the lower temperature also helps in preventing detonation, or knock. Charge boosting is useful only under wide open throttle conditions in s.i. engines, which occur rarely in normal driving, so that such devices are usually used in high-perfor-

512

Internal Combustion Engine Vehicles

boost and removes the need to cap boost pressure under most operating conditions. Turbocharged c.i. engines offer up to 50% higher specic power and torque, and about 10% better fuel economy than naturally aspirated c.i. engines of approximately equal torque capability.

7.2 Superchargers
Most s.i. engine superchargers are driven off the crankshaft and are of the Roots blower or positive displacement pump type. In comparison to turbochargers, these superchargers are bulky and weigh considerably more. In addition, the superchargers are driven off the crankshaft, absorbing 3 to 5% of the engine power output depending on pressure boost and engine speed. The supercharger, however, does not have the low RPM boost problems associated with turbochargers and can be designed to nearly eliminate any time lag in delivering the full boost level. As a result, superchargers are more acceptable to consumers from a driveability viewpoint. The need to reduce engine compression ratio and the superchargers drive power requirement detract from overall efciency. In automotive applications, a supercharged engine can replace a naturally aspirated engine that is 30 to 35% larger in displacement, with a net pumping loss reduction. Overall, fuel economy improves by about 8% or less, if the added weight effects are included. Superchargers are less efcient in combination with c.i. engines, since these engines run lean even at full load, and the power required for compressing air is proportionally greater. Supercharged c.i. engines are not yet commercially available, since the turbocharger appears far more suitable in these application.

8. ALTERNATIVE HEAT ENGINES

A number of alternative engines types have been researched for use in passenger cars but have not yet proved successful in the market place. A brief discussion of the suitability of four engines for automotive power plants is provided next. The Wankel engine is the most successful of the four engines in that it has been in commercial production in limited volume since the 1970s. The thermodynamic cycle is identical to that of a fourstroke engine, but the engine does not use a reciprocating piston in a cylinder. Rather a triangular rotor spins eccentrically inside a Fig. 8shaped

casing. The volume trapped between the two rotor edges and the casing varies with rotor position, so that the intake, comparison, expansion, and exhaust stroke occur as the rotor spins through one revolution. The engine is very compact relative to a piston s.i. engine of equal power, and the lack of reciprocating parts provides smooth operation. However, the friction associated with the rotor seals is high, and the engine also suffers from more heat losses than an s.i. engine. For these reasons, the Wankel engines efciency has always been below that of a modern s.i. piston engine. The two-stroke engine is widely used in small motorcycles but was thought to be too inefcient and polluting for use in passenger cars. One development is the use of direct injection stratied charge (DISC) combustion with this type of engine. One of the major problems with the two-stroke engine is that the intake stroke overlaps with the exhaust stroke resulting in some intake mixture passing uncombusted into the exhaust. The use of a DISC design avoids this problem since only air is inducted during intake. Advanced fuel injection systems have been developed to provide a nely atomized mist of fuel just prior to spark initiation and to sustain combustion at light loads. The two-stroke engines of this type are thermodynamically less efcient than fourstroke DISC engines, but the internal friction loss and weight of two-stroke engine is much lower than a four-stroke engine of equal power. As a result, the engine may provide fuel economy equal or superior to that of a DISC (four-stroke) engine when installed in a vehicle. Experimental prototypes have achieved good results, but the durability and emissions performance of advanced two-stroke engines is still not established. Gas turbine engines are widely used to power aircraft, and considerable research has been completed to assess its use in automobiles. Such engines use continuous combustion of fuel, which holds the potential for low emissions and multifuel capability. The efciency of the engine is directly proportional to the combustion temperature of the fuel, which has been constrained to 12001C by the metals used to fabricate turbine blades. The use of high-temperature ceramic materials for turbine blades coupled with the use of regenerative exhaust waste heat recovery were expected to increase the efciency of gas turbine engines to levels signicantly higher than the efciency of s.i. engines. In reality, such goals have not yet been attained partly because the gas turbine components become less aerodynamically efcient at the small engine

Internal Combustion Engine Vehicles

513

sizes suitable for passenger car use. Part load efciency is a major problem for gas turbines due to the nonlinear efciency changes with airow rates in turbomachinery. In addition, the inertia of the gas turbine makes it poorly suited to vehicle applications, where speed and load uctuations are rapid in city driving. As a result, there is little optimism that the gas turbine powered vehicle will be a reality in the foreseeable future. Stirling engines have held a particular fascination for researchers since the cycle closely approximates the ideal Carnot cycle, which extracts the maximum amount of work theoretically possible from a heat source. This engine is also a continuous combustion engine like the gas turbine engine. While the engine uses a piston to convert heat energy to work, the working uid is enclosed and heat is conducted in and out of the working uid by heat exchangers. To maximize efciency, the working uid is a gas of low molecular weight like hydrogen or helium. Prototype designs of the Stirling engine have not yet attained efciency goals and have had other problems, such as the containment of the working uid. The Stirling engine is, like the gas turbine, not well suited to applications where the load and speed change rapidly, and much of the interest in this engine has faded.

SEE ALSO THE FOLLOWING ARTICLES

Alternative Transportation Fuels: Contemporary Case Studies  Combustion and Thermochemistry  Fuel Cycle Analysis of Conventional and Alternative Fuel Vehicles  Fuel Economy Initiatives: International Comparisons  Hybrid Electric Vehicles  Internal Combustion (Gasoline and Diesel) Engines  Transportation Fuel Alternatives for Highway Vehicles  Vehicles and Their Powerplants: Energy Use and Efciency

Further Reading
Amman, C. (1989). The Automotive EngineA Future Perspective. GM Research Publication GMR-6653. Lichty, C. (1967). Combustion Engine Processes. John Wiley & Sons. National Academy of Sciences (2001). Effectiveness and Impact of Corporate Average Fuel Economy Standards. National Academy Press. Ofce of Technology Assessment (1995). Advanced Automotive TechnologyVisions of a SuperEfcient Family Car. Report to the U.S. Congress, OTA-ETI-638. Weiss, M.A.(Ed.) (2000). On the Road in 2020. Massachusetts Institute of Technology Report MIT-EC00-003.