4
Spark-Ignition Gasoline Engines

INTRODUCTION

A large majority of light-duty vehicles in the United States are powered with spark-ignition (SI) engines fueled with gasoline. Several technologies have been developed to improve the efficiency of SI engines. This chapter updates the status of various SI engine technologies described in the National Research Council report that focused on reduction of fuel consumption (NRC, 2002). As stated in Chapter 2 of the present report, the objective is to evaluate technologies that reduce fuel consumption without significantly reducing customer satisfaction—therefore, power and acceleration performance are not to be degraded. The primary focus is on technologies that can be feasibly implemented over the period to 2025.

The present study examines these SI engine technologies in the context of their incremental improvements in reducing fuel consumption, as well as the associated costs of their implementation. It also discusses the mechanisms by which fuel consumption benefits are realized along with the interactions that these technologies have with the base-engine architecture. As with the other vehicle technologies examined in this report, the committee’s estimates of incremental reduction of fuel consumption and the costs of doing so for the SI technologies presented in this chapter are based on published data from technical journals and analyses conducted by Northeast States Center for a Clean Air Future (NESCCAF), Energy and Environmental Analysis, Inc. (EEA), U.S. National Highway Traffic Safety Administration (NHTSA), U.S. Environmental Protection Agency (EPA), and other organizations. In addition, the expert judgment of committee members whose careers have focused on vehicle and power train design, development, and analysis, as well as the results of consultation with individual original equipment manufacturers (OEMs) and suppliers, were also incorporated in the estimates.

SI ENGINE EFFICIENCY FUNDAMENTALS

It is common practice to group engine-efficiency-related factors with their respective process fundamentals (i.e., thermodynamic factors, friction losses, etc.). For example, consider the basic stages of the SI engine cycle that contribute to positive work: heat released during fuel combustion, volumetric expansion, and associated heat transfer. The factors related to this process can be grouped together as the thermodynamic component. In addition, there are several processes within the engine that mitigate the positive work produced; these can be grouped as either gas exchange losses (pumping losses) or frictional losses within the engine. Further more, the engine architecture and the use of accessory/operational components (i.e., power steering, coolant, oil and fuel pumps) can be the source of additional parasitic losses. The fundamental aspects of each category of engine efficiency factors are discussed further in the following sections.

Thermodynamic Components

Thermodynamic factors include combustion interval, effective expansion ratio, and working fluid properties. In consideration of these factors there are some fundamental methods that can be used to improve efficiency, including:

  • Short combustion intervals—allow for more of the heat of combustion to undergo more expansion and thus yield an increase in positive work.

  • High compression ratios and late exhaust-valve-opening event—can be used to influence the expansion ratio in order to improve efficiency. However, these factors are constrained by other considerations.

  • High specific heat ratio of working fluid (i.e., cp/cv.)—working-fluid property of significance related to the specific heat ratio. Atmospheric air is preferred over exhaust gas as a combustion diluent thermodynamically, but exhaust emis-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 38
4 Spark-Ignition Gasoline Engines INTRODUCTION SI ENGINE EFFICIENCY FUNDAMENTALS A large majority of light-duty vehicles in the United It is common practice to group engine-efficiency-related States are powered with spark-ignition (SI) engines fueled factors with their respective process fundamentals (i.e., with gasoline. Several technologies have been developed to thermodynamic factors, friction losses, etc.). For example, improve the efficiency of SI engines. This chapter updates consider the basic stages of the SI engine cycle that contrib- the status of various SI engine technologies described in the ute to positive work: heat released during fuel combustion, National Research Council report that focused on reduction volumetric expansion, and associated heat transfer. The fac- of fuel consumption (NRC, 2002). As stated in Chapter 2 of tors related to this process can be grouped together as the the present report, the objective is to evaluate technologies thermodynamic component. In addition, there are several that reduce fuel consumption without significantly reducing processes within the engine that mitigate the positive work customer satisfaction—therefore, power and acceleration produced; these can be grouped as either gas exchange performance are not to be degraded. The primary focus is losses (pumping losses) or frictional losses within the en - on technologies that can be feasibly implemented over the gine. Furthermore, the engine architecture and the use of period to 2025. accessory/operational components (i.e., power steering, The present study examines these SI engine technolo- coolant, oil and fuel pumps) can be the source of additional gies in the context of their incremental improvements in parasitic losses. The fundamental aspects of each category of reducing fuel consumption, as well as the associated costs engine efficiency factors are discussed further in the follow- of their implementation. It also discusses the mechanisms ing sections. by which fuel consumption benefits are realized along with the interactions that these technologies have with the base- Thermodynamic Components engine architecture. As with the other vehicle technologies examined in this report, the committee’s estimates of in- Thermodynamic factors include combustion interval, cremental reduction of fuel consumption and the costs of effective expansion ratio, and working fluid properties. In doing so for the SI technologies presented in this chapter are consideration of these factors there are some fundamental based on published data from technical journals and analyses methods that can be used to improve efficiency, including: conducted by Northeast States Center for a Clean Air Future (NESCCAF), Energy and Environmental Analysis, Inc. • Short combustion intervals—allow for more of the heat (EEA), U.S. National Highway Traffic Safety Administration of combustion to undergo more expansion and thus yield an (NHTSA), U.S. Environmental Protection Agency (EPA), increase in positive work. and other organizations. In addition, the expert judgment of • High compression ratios and late exhaust-valve- committee members whose careers have focused on vehicle opening event—can be used to influence the expansion ratio and power train design, development, and analysis, as well as in order to improve efficiency. However, these factors are the results of consultation with individual original equipment constrained by other considerations. manufacturers (OEMs) and suppliers, were also incorporated • High specific heat ratio of working fluid (i.e., cp/cv.)— in the estimates. working-fluid property of significance related to the specific heat ratio. Atmospheric air is preferred over exhaust gas as a combustion diluent thermodynamically, but exhaust emis- 38

OCR for page 38
39 SPARK-IGNITION GASOLINE ENGINES Engine Architecture sions after-treatment challenges limit this as an option for reducing fuel consumption. Engine architecture refers to the overall design of the en- • Optimize timing of spark event—an important factor gine, generally in terms of number of cylinders and cylinder since this affects the countervailing variables of in-cylinder displacement. The engine architecture can affect efficiency heat loss and thermodynamic losses. This is discussed in mainly through bore-stroke ratio effects and balance-shaft more detail below. requirements. Trends in power train packaging and power-to-weight Maximum efficiency occurs when the two countervail- ratios have led in-line engines to have under-square bore- ing variables, heat loss and thermodynamic losses, sum to stroke ratios (i.e., less than unity) while most V-configuration a minimum. The optimum spark timing is often referred engines have over-square ratios. Under-square ratios tend to to as minimum advance for best torque or maximum brake be favored for their high thermodynamic efficiency. This is torque (MBT). At low to moderate speeds and medium to due to the surface-area-to-volume ratio of the combustion high loads, SI engines tend to be knock-prone, and spark- chamber; under-square designs tend to exhibit less heat timing retardation is used to suppress the knock tendency. transfer and have shorter burn intervals. Over-square designs Spark-timing adjustments are also made to enable rapid- enable larger valve flow areas normalized to displacement response idle load control to compensate for such things as and therefore favor power density. These interactive factors AC compressor engagement. For this to be effective, idle play a role in determining overall vehicle fuel efficiency. spark timing must be substantially retarded from MBT. Re- Balance-shafts are used to satisfy vibration concerns. tardation from MBT for either of the aforementioned reasons These balance shafts add parasitic losses, weight, and ro- compromises fuel consumption. tational inertia, and therefore have an effect on vehicle fuel efficiency. I4 engines having displacement of roughly 1.8 L Gas Exchange or Pumping Losses or more require balance shafts to cancel the second-order shake forces. These are two counter-rotating balance shafts Gas exchange or pumping losses, in the simplest terms, running at twice crankshaft speed. The 90° V6 engines typi- refer to the pressure-gradient-induced forces across the cally require a single, first-order balance shaft to cancel a piston crown that oppose normal piston travel during the rotating couple. The 60° V6 and 90° V8 engines need no exhaust and intake strokes. The pumping loss that princi- balance shafts. Small-displacement I3 engines have received pally affects fuel consumption is that which occurs during development attention from many vehicle manufacturers. the intake stroke when the cylinder pressure and the intake These require a single first-order balance shaft to negate manifold are approximately equal. The pumping loss compo- a rotating couple. While low-speed high-load operation of nent that occurs during the exhaust stroke mainly affects peak small displacement I3 engines tends to be objectionable power. Both of these oppose the desired work production of from a noise, vibration, and harshness (NVH) perspective, the engine cycle and thus are seen as internal parasitic losses, they could be seen as candidate engines for vehicles such as which compromise fuel efficiency. hybrid-electric vehicles (HEVs) where some of the objec- tionable operating modes could be avoided. Frictional Losses Parasitic Losses The main source of friction losses within an SI engine are the piston and crankshaft-bearing assemblies. The majority Parasitic losses in and around the engine typically involve of the piston-assembly friction comes from the ring-cylinder oil and coolant pumps, power steering, alternator, and bal- interface. The oil-control ring applies force against the cylin- ance shafts. These impose power demands and therefore der liner during all four strokes while the compression rings affect fuel consumption. Many vehicle manufacturers have only apply minor spring force but are gas-pressure loaded. given much attention to replacing the mechanical drives for Piston-assembly friction is rather complex as it constantly the first three of these with electric drives. Most agree that undergoes transitions from hydrodynamic to boundary-layer electrification of the power steering provides a measurable friction. Hydrodynamic piston-assembly friction predomi- fuel consumption benefit under typical driving conditions. nates in the mid-stroke region while boundary-layer friction Fuel consumption benefit associated with the electrification is common near the top center. Avoidance of cylinder out- of oil or coolant pumps is much less clear. Electrification of of-roundness can contribute to the minimization of piston- these functions provides control flexibility but at a lower effi- ring-related friction. Crankshaft-bearing friction, while ciency. Claims have been made that the coolant pump can be significant, is predominately hydrodynamic and is relatively inactive during the cold-start and warm-up period; however, predictable. consideration must be given to such things as gasket failure, bore or valve seat distortion, etc. These factors result from

OCR for page 38
40 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES local hot spots in the cooling system since much of the waste may add another 1 to 2 percent benefit at a cost of $50, $80, heat enters the cooling system via the exhaust ports. and $100 for I4, V6, and V8 engines, respectively, based on Further discussion on the parasitic losses associated with manufacturer’s input. As of 2007 the implementation of this these types of engine components is provided in Chapter 7 technology has become common; therefore, fast burn and of this report. strategic EGR is considered to be included in the baseline of this analysis. THERMODYNAMIC FACTORS Variable Compression Ratio Fast-Burn Combustion Systems If an engine’s compression ratio could be adjusted to near Fast-burn combustion systems are used to increase the knock-limited value over the operating range, significant the thermodynamic efficiency of an SI engine by reduc - fuel economy gains could be realized. Many mechanisms to ing the burn interval. This is generally achieved either realize variable compression ratios have been proposed in by inducing increased turbulent flow in the combustion the literature and many have been tested. However, to date chamber or by adding multiple spark plugs to achieve rapid all these attempts add too much weight, friction, and para- combustion. sitic load as well as significant cost and have therefore not Fluid-mechanical manipulation is used to increase turbu- been implemented into production designs (Wirbeleit et al., lence through the creation of large-scale in-cylinder flows 1990; Pischinger et al., 2001; Tanaka et al., 2007). It should (swirl or tumble) during the intake stroke. The in-cylinder be recalled that alterations to the effective compression ratio flows are then forced to undergo fluid-motion length-scale via intake-valve closing (IVC) timing adjustments with reduction near the end of the compression stroke due to the higher-than-normal geometric compression ratios achieves reduced clearance between the piston and the cylinder head. some of this benefit. This reduction cascades the large-scale fluid motion into smaller scale motions, which increases turbulence. Increased VALVE-EVENT MODULATION OF GAS-EXCHANGE turbulence increases the turbulent flame speed, which there- PROCESSES by increases the thermodynamic efficiency by allowing for reduced burn intervals and by enabling an increase in knock- Alteration of valve timing can have a major impact on limited compression ratio by 0.5 to 1.0. This decrease in volumetric efficiency over an engine’s speed range, and burn interval increases dilution tolerance of the combustion thus peak torque and power are affected by this. IVC timing system. Dilution tolerance is a measure of the ability of the is the main determinant of this effect (Tuttle, 1980). Early combustion system to absorb gaseous diluents like exhaust IVC (compression stroke) favors torque, and later IVC gas. Exhaust gas is introduced by means of an exhaust-gas- favors power. Implementations of valve-event modulation recirculation (EGR) system or by a variable-valve-timing (VEM) typically are referred to as specific technologies scheme that modulates exhaust-gas retention without incur- such as variable valve timing, variable valve timing and ring unacceptable increases in combustion variability on a lift, two-step cam phasing, three-step cam phasing, and cycle-by-cycle basis. Combustion variability must be con- intake-valve throttling. VEM aids fuel consumption reduc - trolled to yield acceptable drivability and exhaust emissions tion by means of reducing pumping loss. Pumping loss is performance. reduced by either allowing a portion of the fresh charge to Multiple spark plugs are sometimes used to achieve rapid be pushed back into the intake system (late IVC during the combustion where fluid-mechanical means are impractical. compression stroke) or by allowing only a small amount Here, multiple flame fronts shorten the flame propagation of the mixture to enter the cylinder (early IVC during the distance and thus reduce the burn interval. High dilution- intake stroke). tolerant combustion systems can accept large dosages of It should be noted that any of the VEM schemes that EGR, thereby reducing pumping losses while maintaining reduce or eliminate the pumping loss also reduce or elimi- thermodynamic efficiency at acceptable levels. nate intake-manifold vacuum. Alternative means to oper- ate power brakes, fuel vapor canister purge, and positive c rankcase ventilation (PCV) systems, normally driven Fuel Consumption Benefit and Cost of Fast-Burn by intake-manifold vacuum, must then be considered. To Combustion Systems overcome this issue, an electrically operated pump may Combining fast-burn and strategic EGR usage typically need to be added. It should also be noted that while the decreases fuel consumption by 2 to 3 percent, based on implementation of VEM techniques can boost torque output manufacturer’s input. The implementation of this technology of a given engine, this report assumes that constant torque is essentially cost neutral. Variable mixture-motion devices, will be maintained, leading to engine downsizing. The fuel which may throttle one inlet port in a four-valve engine to consumption benefits listed in the following section consider increase inlet swirl and in-cylinder mixture momentum, a constant-torque engine.

OCR for page 38
41 SPARK-IGNITION GASOLINE ENGINES VEM History tive expansion ratio. To achieve a lower effective compres- sion ratio, the intake valve closing is delayed until later on The first modern successful production implementation of the compression stroke at light loads. By closing the valve a varying valve-event setup was Honda’s VTEC in the late later on the compression stroke, a larger portion of the air 1980s. Honda’s system allowed a stepped increase in the that was drawn in on the intake stroke is pushed back out duration and lift of the intake valves. Prior to the develop- through the valve. This phenomenon allows a decrease in ment of a multi-step cam profile system, a cam profile was pumping losses by relying on the timing of the intake valve chosen based on performance compromises. Engineers were to regulate engine load. From the reduction in pumping confronted with a tradeoff, as it is difficult to satisfy the needs losses, a reduction in fuel consumption will occur. Some of both good low-speed torque and high-speed torque with a refer to late IVC as the Atkinson cycle (Boggs et al., 1995), single cam profile. The cam profiles and timings necessary and most engines have some of this character. For boosted to maximize these needs are completely different in their engines, late IVC is termed by some as the Miller cycle characteristics. (Hitomi et al., 1995). Honda’s technology was one of the first discrete vari- A diagram of a typical oil-actuated variable cam phaser able valve lift (DVVL)-type systems. Over the years, many system installed on the intake cam (exhaust cam timing for other companies have developed various implementations this engine is fixed), Figure 4.1 shows the complexity of of DVVL-type setups, as well as other innovative VEM integrating a variable cam phaser into the standard engine technologies. Some newer developments in VEM tech - architecture with fixed timing. As indicated in the figure, nology include systems that offer continuously variable two separate oil passages are fed to the phaser. A solenoid lift and duration. Nissan’s VEL, BMW’s Valvetronic, and controls the direction of the fluid to the two different pas- Fiat’s Multi-Air are all examples of continuously variable sages. These passages are used to control whether the cam lift systems that also incorporate adjustable valve timing will be advanced or retarded relative to the crankshaft. In (Takemura et al., 2001; Flierl and Kluting, 2000; Bernard et order for the engine control unit (ECU) to sense the relative al., 2002). These systems attempt to operate throttle-less and position of the camshaft, a position sensor is installed that rely on varying lift and timing to throttle the incoming air. provides feedback information to the ECU. It is important to Throttle-less operation allows a reduction in pumping losses note that, like many of the vehicle technologies discussed in at part load, and thus reduces fuel consumption. However, this chapter, implementing a variable cam phaser involves a these throttle-less approaches also generally result in slight complete system integration as illustrated in Figure 4.1 and variations in the very small valve lifts necessary for idle is not as simple as bolting on a component. operation even with well-controlled manufacturing toler- ances. These small variations result in a slightly different Fuel Consumption Benefit and Cost of IVC Timing charge mass from cylinder to cylinder, causing somewhat rougher idle engine operation, which is detrimental to cus - OEM input suggests intake cam phasing results in roughly tomer satisfaction. a 1 to 2 percent fuel consumption reduction. Both the EPA The cam phaser, used to vary the valve timing, is another and NESCCAF also estimate approximately 1 to 2 percent technology that has been in constant development by the fuel consumption reduction (EPA, 2008; NESCCAF, 2004). OEMs. Early cam phasers featured only two-step phasing, al- EEA claims a fuel consumption improvement of 1.1 to 1.7 lowing two possible cam positions relative to the crankshaft. percent can occur with the addition of an ICP (EEA, 2007). Today, cam phasing is fully variable, offering a wide range In agreement with most sources, the committee has also es- of positions. Due to the system’s relative simplicity and long timated a 1 to 2 percent reduction in fuel consumption using evolution, many production vehicles now utilize cam phas- ICP. However, as with the other VEM technologies that are ing technology. Until recently, cam phasing had only been listed in the chapter, a generalized statement can be made that applied to overhead cam style setups due to ease of integra- smaller-cylinder-count engines (i.e., four cylinders) will be tion. This recently has changed with GM’s development and closer to the low end of this improvement range, and higher- production of an in-block cam phaser applied to its overhead cylinder-count engines will be closer to the high end of the valve (OHV) 6.2-L engine. fuel consumption reduction ranges that are listed. OEM input suggests that fixed-duration intake systems Intake-Valve Closing Timing add a cost of about $35/phaser. OEM input does not reflect a retail price equivalent (RPE) factor. The EPA estimates an Intake-valve closing timing, also known as intake cam RPE cost increase of $59/phaser (EPA, 2008). NESCCAF phasing (ICP), is a form of VEM. At moderate speeds and quoted a literature RPE of $18 to $70 (NESCCAF, 2004) light loads, late intake valve closing (i.e., during the com - and EEA estimates an RPE of $52/phaser (EEA, 2007). A pression stroke) can reduce the pumping loss; however, it 1.5 RPE factor was used to develop the committee estimate also slows combustion. Typically this configuration yields of $52.50 for an in-line engine and $105 for a V-configuration effective compression ratios that are lower than the effec - that requires two phasers.

OCR for page 38
42 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES FIGURE 4.1 System-level mechanization of the variable cam phaser, oil control valve, control module, crank sensor, and cam sensor to the engine. SOURCE: Delphi (2009). Reprinted by permission from Delphi Corporation. Figure 4-1.eps bitmap Valve Overlap Control Fuel Consumption Benefit and Cost of Valve Overlap Control Valve overlap control, also known as dual cam phasing The fuel consumption reduction from valve overlap (DCP), is another form of VEM. Valve overlap (i.e., the control/DCP is expected to be slightly greater than just interval between intake-valve opening [IVO] and exhaust- controlling the IVC timing at about 2 percent over intake valve closing [EVC]) can affect residual-gas retention at low phasing alone, based on manufacturer input. The EPA and loads and can reduce pumping loss in a manner similar to NESCCAF both estimate a reduction in consumption of 2 that with EGR (exhaust gas recirculation). Valve overlap con- to 4 percent (EPA, 2008; NESCCAF, 2004). EEA estimates trol can also be utilized to tune performance at high engine a 1.8 to 2.6 percent improvement in fuel economy (EEA, speeds, resulting in increased torque, which, in principle, 2007). The committee concluded that adding variable ex- can allow for minor engine downsizing. Valve overlap can haust cam phasing to ICP will yield an incremental 1.5 to 3 be modulated by changing the phasing of either the intake or percent reduction in fuel consumption. This would mean the exhaust cam. Typically it is done with the exhaust cam be- total estimated effect of adding DCP would be about 2.5 to cause exhaust-cam phasing for increased overlap also delays 5 percent over an engine without any variable valve timing exhaust-valve opening (EVO) timing. Thus both EVO and technology. The high end of 5 percent has been verified by EVC move in ways favorable to low-speed and light-load OEMs and Ricardo, Inc.’s full-vehicle system simulation fuel consumption reduction. Modulating valve overlap with (FSS) (Ricardo, Inc., 2008). an intake cam yields countervailing effects, i.e., increased Dual overhead cam (DOHC) V-engines with variable valve overlap in this manner tends to reduce pumping loss intake and exhaust would require four cam phasers, adding while the corresponding IVC event will occur earlier, thus roughly $140 of manufacturer cost based on manufacturer offsetting some of the increased-overlap benefit. At idle, input, but a portion of this is offset by the elimination of the too much valve overlap will destabilize combustion. When external EGR system. EEA estimates an RPE of $76 to $84 variable phasing, fixed-duration intake and exhaust cams are for an I4, and $178 to $190 for V6 and V8 engines (EEA, implemented, valve-overlap control may eliminate the need 2007). The EPA estimates an incremental cost increase of for an external EGR system. $89 for an I4 and $209 for V6 and V8 engines (EPA, 2008).

OCR for page 38
43 SPARK-IGNITION GASOLINE ENGINES NESCCAF quotes a literature RPE of $35 to $140 for dual VTEC system is more cost-effective on its single overhead cam phasers (NESCCAF, 2004). Discussion with OEMs also cam (SOHC) engines, due simply to the fact that a DOHC verified that by simply doubling the cost of ICP, a reason- engine would require more hardware. This is an example ably accurate DCP cost can be attained. The committee has of one manufacturer’s method of DVVL implementation. estimated an RPE cost of $52.50 for an in-line engine and It should be noted that other manufacturers have developed $105 for a V-configuration, incremental to the cost of ICP different designs to accomplish the same goal, and as a result technology. the different systems have differing amounts of pumping loss reduction and friction increase. This situation reinforces the point that advanced VEM technologies are not simply “bolt- Intake-Valve Throttling on” parts that provide a uniform fuel consumption reduction Using very short duration and low-lift intake-valve- to all OEMs. opening events during the intake stroke can reduce (or Delphi performed testing on a GM 4.2-L I6 equipped eliminate) the pumping loss. This VEM, also known as with a two-step variable valve actuation system and a cam- intake-valve throttling, also tends to slow combustion, shaft phaser on the intake (Sellnau et al., 2006). The engine mainly at low engine speeds. (Small-scale turbulence gener- was already outfitted with an exhaust cam phaser. Delphi’s ated by this approach dissipates rapidly, well before the start two-step valve actuation system consisted of oil-actuated of combustion, and thus this does not generally contribute switchable rocker arms. Testing on the engine revealed a to rapid combustion). Note that low valve lift is simply a 4.3 percent fuel consumption reduction during the EPA city consequence of short-duration cam design. Manufacturing drive cycle, compared to the base engine with no variable tolerance control is of extreme importance with intake valve lift and timing. These results were obtained with no other throttling if cylinder-to-cylinder variability at idle is to be modifications besides the VVL, a phaser, and control system acceptable. BMW and Nissan currently offer this technol- reconfiguration. Delphi claimed that “mixture motion is ogy on some of their engine models, which use varying lift nearly absent for low lift profiles, so an enhanced combus- and timing to throttle the engine. Other manufacturers have tion system, with higher tumble for low-lift profiles, would announced plans to introduce engines with throttle-less op- likely yield significant improvements in fuel economy.” In eration within the next few years. the second portion of the test Delphi modified the cylinder The above options (DCP and ICP) are focused mainly on head and added flow restriction that generates turbulence in pumping-loss reduction by means of late IVC timing and an attempt to speed up combustion, thereby furthering the exhaust-gas recycling via variable valve overlap. Very early fuel economy gain. Chamber masks were used to increase IVC (i.e., during the intake stroke) is another effective means the tumble motion. The lift profile on the exhaust cam and the of reducing pumping losses, but it involves much more port were also modified. For the second phase of testing with complex and costly means of implementation. Two types the altered cylinder head and calibration, the fuel consump- of intake-valve-opening techniques are considered: discrete tion reduction was estimated to be 6.5 percent in comparison variable valve lift and continuously variable valve lift. to the original engine. These values were estimated from data taken at multiple load points rather than over a driving cycle (Sellnau et al., 2006). Discrete Variable Valve Lift A discrete variable valve lift (DVVL) system is one Fuel Consumption Reduction and Cost of DVVL which typically uses two or three different cam profiles over the range of engine speeds and loads. This system attempts Two (or three)-step cams that yield short intake durations to reduce pumping losses by varying the lift profile of the using DVVL can yield fuel consumption reductions in the camshaft. By varying the lift of the valves, it is possible 4 to 5 percent range based on vehicle OEM input. A reduc- to limit the use of the throttle and significantly reduce the tion of 3 to 4 percent in fuel consumption (FC) is estimated pumping losses. from the EPA (EPA, 2008). FEV has developed a two-stage As described earlier, Honda has been using a DVVL-type switch of the intake valve lift that is claimed to offer up to a setup on its vehicles known as VTEC. To engage the differ- 6 to 8 percent reduction in consumption when combined with ent cam profile on Honda’s system, there is a third cam lobe variable valve timing, during the New European Drive Cycle and follower, located in between the two main lobes, which (Ademes et al., 2005). NESCCAF and EEA estimate that a is hydraulically activated by an internal solenoid controlled 3 to 4 percent reduction is possible (NESCCAF, 2004; EEA, oil passage. During low-speed and low-load operation, the 2007) on the U.S. driving cycles. EEA also estimates a fuel engine runs using the base cam profile(s). Once a certain economy improvement of 7.4 to 8.8 percent when DVVL is load point is reached, the ECU activates a control valve to combined with DCP and the engine is downsized to maintain direct oil pressure from the main gallery to an oil passage constant torque. Simulation work by Sierra Research indi- that engages the third follower. Once the third follower en- cated a 6.3 to 6.8 percent benefit when combined with vari- gages, it is then locked into place by a locking pin. Honda’s able valve timing, which accounts for up to 5 percent of that

OCR for page 38
44 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES amount (Sierra Research, 2008). The committee concluded that a 1.5 to 4.0 percent drive-cycle-based FC reduction is possible, incremental to an OHC engine with DCP or an OHV engine with CCP. Vehicle OEM input suggests a $35 to $40/cylinder cost for implementing DVVL. The Martec Group estimates an OEM cost of $320 to implement a two-step VVL on a V6 DOHC engine (Martec Group, Inc., 2008). The EPA estimates an incremental cost increase of $169 for an I4, $246 for a V6, and $322 for a V8 (EPA, 2008). EEA estimates RPEs for an OHC-4V; $142 to $158 (equivalent to $95 to $105 assuming an RPE multiplier of 1.5) for an I4, $188 to $212 (equivalent to $125 to $141 assuming an RPE multiplier of 1.5) for a V6, and $255 to $285 (equivalent to $170 to $190 assuming an FIGURE 4.3 Nissan valve event and lift design. SOURCE: RPE multiplier of 1.5) for a V8 (EEA, 2007). The committee Takemura et al. (2001).Figure 4-3.eps Reprinted with permission from SAE Paper estimates the manufacturing cost of implementing DVVL to 2001-01-0243, Copyright 2001 SAE International. bitmap be about $30 to $40/cylinder. Continuously Variable Valve Lift to its relative novelty to the mass production environment The continuously variable valve lift (CVVL) system and the large fuel consumption benefits it offers. Two ap- allows a wide control range of the camshaft profile (see proaches to CVVL have been considered, electromechanical Figures 4.2 and 4.3 for schematics). A continuous system and electrohydraulic. allows for calibration of the optimal valve lift for various load conditions, versus the discrete system, which will only offer Electromechanical CVVL Systems two or three different profiles. The combination of a continu- ous VVL system and an intake cam phaser has the potential BMW was the first to offer a mass production fully vari- to allow the engine to operate throttle-less. In the following, able valve train incorporating CVVL in 2001, which it calls greater detail of this particular VEM technology is given due Valvetronic, Figure 4.2. This system is an electromechanical system that when combined with variable intake and exhaust cam phasers provides a fully throttle-less induction system. To vary the lift of the valve, an intermediate lever was added along with an eccentric shaft. The eccentric shaft is operated by an electric motor that adjusts the positioning of the lever over the camshaft. The lever contains a profile with one side being relatively flat and the other side being relatively steep. Adjusting the relative positioning of the lever controls the valve lift. BMW claims that up to a 10 percent reduction in fuel consumption is possible with this system (Sycomoreen). Figure 4.2 shows the many added components needed for the Valvetronic system. Nissan Motor Company has also developed a continuous variable valve event and lift (VEL) system (Figure 4.3). The electromechanical system allows continuous variation of valve timing and lift events similar to the BMW system, but achieves this using a different architecture. Nissan estimates a 10 percent reduction in fuel consumption over the Japanese 10-15 drive cycle (Takemura et al., 2001) for its VEL system. The 10-15 drive cycle is intended to simulate a typical urban drive cycle, and an EPA combined FTP cycle rating would be somewhat lower. Nissan attributes the reduction in con- sumption to “lower friction loss due to the use of extremely small valve lift-timing events and reduction of pumping loss FIGURE 4.2 BMW Valvetronic. SOURCE: Flierl et al. (2006). resulting from effective use of internal gas recirculation.” Reprinted Reprinted with permission from SAE Paper 2006-01-0223, Copy- Copy- Nissan evaluated the consumption benefits distribution at a right 2006 SAE International. 4-2.eps Figure bitmap

OCR for page 38
45 SPARK-IGNITION GASOLINE ENGINES fixed speed and load of 1,600 rpm and 78 N-m. The distribu- tion of effects was the following: (1) pumping loss decrease yielded a consumption reduction of 5.2 percent, (2) friction reduction yielded a consumption benefit of 1.1 percent, and (3) an improvement in combustion caused a reduction in consumption of 3 percent. Figure 4.3 shows the layout of Nissan’s VEL system. The electromechanical system uses an oscillating cam to open and close the valve. An oscillating cam (output cam) looks like half of a camshaft, but it is hinged on one end to allow full opening and closing of the valve on the same cam face. To change the valve lift and duration of the cam, the control shaft is adjusted by a motor to change the distance between the control cam and the oscillating cam. An increase in dis- tance is caused by the lobe on the control shaft turning and pushing the rocker arm assembly out. This changes which portion of the output cam contacts the valve to control the FIGURE 4.4 Univalve. SOURCE: Flierl et al. (2006). Reprinted amount of lift. with permission fromFigure 4-4.eps SAE Paper 2006-01-0223, Copyright 2006 Toyota Motor Company has recently developed its own SAE International. bitmap type of a CVVL timing system. The new system will first be applied to their newly developed 2.0-L engine. Toyota’s system features separate cam phasers on the intake and ex- haust camshafts to vary the camshaft timing, along with a production and the testing cycle used to produce this estimate continuously variable valve lift system. Toyota claims that is unclear. Therefore, Advanced-VTEC is only mentioned the system “improves fuel efficiency by 5 to 10 percent to demonstrate an example of emerging CVVL technology. (depending on driving conditions), boosts output by at least 10 percent and enhances acceleration.” Toyota did not state Electrohydraulic CVVL Systems what features the base engine already had in order to gener- ate fuel efficiency improvement percentages (Toyota Motor The electrohydraulic approach to CVVL has been under Co., 2007). development for over a decade. One of the organizations The Technical University of Kaiserslautern performed which has been active in this development is Fiat Central testing on a 2.0-L four-cylinder gasoline engine that was Research (CRF). The major focus of the work by CRF is outfitted with a fully variable lift and timing system (VVTL) a system that it calls Uniair (Bernard et al., 2002). Fiat re- called Univalve, Figure 4.4. The Univalve system allows for cently announced a system it calls Multiair that is derived either the use of standard throttle or unthrottled operation. At from Uniair. Multiair is a joint development between Fiat a load point of 2000 rpm and a BMEP of 2 bar, a 13 percent and valve train component supplier INA that promises a 10 reduction in fuel consumption occurred compared to the base percent reduction in fuel consumption. Other organizations engine with a nonvariable valve train. This reduction is due have also been active in the development of systems using to the reduction in the pumping work and an improvement in similar principles (Misovec et al., 1999). The Uniair/Multiair the formation of the mixture. The Univalve system varies the system has been described as a lost-motion system wherein lift and duration of the valve by adjusting the eccentric con- the camshaft lobe drives the piston of a small pumping cham- tour (see Figure 4.4). Adjusting the eccentric shaft changes ber, one for each cylinder intake and one for each exhaust. the rocker arm pivot point (Flierl et al., 2006). Multiair utilizes the system only for the intake valves. The Univalve system in Figure 4.4 operates similar to The output from the pump is controlled by a solenoid- BMW’s version of a CVVL system. In Figure 4.4 the image actuated flow control valve that directs the hydraulic output to the left demonstrates a fixed pivot ratio on the rocker with of the pump directly to the hydraulic actuator on the valve(s) constant valve lift. The image to the right features variable or to the accumulator. If the control valve directs the hy- valve lift. To vary the lift the rocker arm is no longer fixed draulic pressure to the valve actuator(s), the valve(s) open to a single pivot point. An eccentric shaft creates a varying normally following the camshaft profile. In principle a lost- pivot point by adjustment of the shaft’s contour contact point motion system allows opening the valve(s) at any fraction of on the rocker. the normal valve lift profile by directing part of the hydraulic Honda has also patented its new Advanced-VTEC system, pressure to the accumulator rather than to the valve actuator. which turns its current DVVL VTEC system into a throttle- By appropriately controlling the application of the hydraulic less CVVL setup. While initial claims are up to a 10.5 percent pressure to the valve actuators or to the accumulator, a wide reduction in fuel consumption, this system is not currently in range of valve lift profiles can be achieved, including mul-

OCR for page 38
46 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES VEM Implementation Techniques tiple small lifts during one valve event. This latter capability is not achievable with mechanical CVVL systems. However, Many of the above-mentioned VEM systems are often electrohydraulic CVVL systems tend to be less efficient implemented as a package combining varying valve lift and considering the energy lost by the hydraulic pump and the timing events. The combination of these technologies will increased friction losses from the additional number of com- provide further reduction in the use of the throttle. ponents. The committee believes that the large increase in General Motors Research and Development (Kuwahara parasitic losses that will offset the perceived fuel consump- et al., 2000; Cleary and Silvas, 2007) performed testing tion reduction benefit, combined with the high component on a single-cylinder model of their 3.4-L DOHC engine. cost, will limit the market penetration of this technology. In The model made use of varying intake valve cam timing, addition, achieving consistent and uniform valve lifts under duration, and intake valve lift. A combination of the vary- idle conditions to maintain a smooth idle may be more chal- ing parameters allowed for the engine to operate without a lenging than with mechanical CVVL systems. throttle. From the study by General Motors, an approximate reduction in fuel consumption of up to 7 percent occurred Fuel Consumption Benefit and Cost of CVVL at part load conditions. By unthrottling the engine, a large reduction in throttling losses occurs and the engine was able The above discussion reviewed the technology of VEM to operate at higher intake manifold pressures. It is important approaches and various FC benefits ascribed to each system. to note that the cost and fuel consumption reductions of the As noted in Chapter 2, the fuel consumption reduction ben- various VEM approaches are highly variable and dependent efits for the technology approaches considered are based on upon the basic engine architecture to which they are applied. the combined city and highway driving cycles, while some of the benefits described earlier are not necessarily based Cylinder Deactivation on these driving cycles. CVVL is expected to be in the 5 to 7 percent range based on manufacturer input. The EPA and Cylinder deactivation is utilized during part load situ- NESCCAF both estimate a 4 to 6 percent reduction in fuel ations to reduce thermal and throttling losses. During consumption (EPA, 2008; NESCCAF, 2004), while EEA constant speed operation, the power demand is relatively estimates a 6.5 to 8.3 percent reduction in fuel consumption low. By shutting off multiple cylinders, a higher load is at constant engine size and 8.1 to 10.1 percent with an engine placed on the remaining operating cylinders. The higher downsize to maintain constant performance (EEA, 2007). load requires the throttle to be open further and therefore Sierra Research’s simulation work resulted in a 10.2 to 11.0 reduces the throttling losses. The decrease in losses reduces percent benefit when combined with variable valve timing the overall fuel consumption. Cylinder deactivation via (Sierra Research, 2008). The committee has estimated that valve deactivation has been applied to four-, six-, and eight- CVVL will have an additional 3.5 to 6.5 percent reduction in cylinder engines, in some cases rather successfully. Most fuel consumption over an engine already equipped with DCP. commonly, cylinder deactivation is applied to engines that Going from a base DOHC engine to one with continuously have at least six cylinders; four-cylinder engines typically are variable lift and timing could provide a 6 to 11 percent fuel not equipped with deactivation due to additional noise, vibra- consumption reduction assuming engine size adjustments for tion, and harshness concerns that are deemed unsatisfactory constant acceleration performance. for consumers. Even current production V6 offerings have Vehicle OEM input suggests that the cost of a continu- NVH levels that are very noticeable to customers. Increased ously variable intake-valve is two to three times that of the NVH can be perceived as a low-quality characteristic that two-step system plus the cost of the actuation system ($40 deters potential customers from purchasing vehicles with to $80) plus the cost of the intake and exhaust cam-phasing this technology. system. Vehicle integration could add another cost in the range of $140. The EPA estimates an RPE incremental cost History of Cylinder Deactivation of $254 (or $169 cost assuming an RPE multiplier of 1.5) for I4, $466 (or $311 cost) for V6, and $508 (or $339 cost) Cylinder deactivation was first implemented on a pro- for V8 engines (EPA, 2008). The Martec Group estimates duction vehicle in 1981 on the Cadillac V8-6-4. The engine a manufacturing cost of $285 for an I4, $450 for a V6, and could operate in four-, six-, and eight-cylinder mode depend- $550 for a V8 (Martec Group, Inc., 2008). For a CVVL sys- ing on power demand. To deactivate the cylinders, a solenoid tem, EEA (2007) estimates RPEs of $314 to $346 (or $209 to mounted on top of the rocker arm assembly would disconnect $231 cost) for an I4, $440 to $480 (or $293 to $320 cost) for the pivot point for the rocker and the rocker would then pivot a V6, and $575 to $625 (or $383 to $417 cost) for a V8 (EEA, against a soft spring. The valves would remain closed and 2007), all assuming an RPE multiplier of 1.5. The commit- the cylinder would not fire, but rather act as a compressed tee estimates the manufacturing cost of CVVL to be $159 to air spring. This system helped to reduce fuel consumption at $205 for I4 engines, $290 to $310 for V6 engines, and $350 cruising type conditions. However, drivability and the need to $390 for V8 engines, not including an RPE factor.

OCR for page 38
47 SPARK-IGNITION GASOLINE ENGINES for quick re-engagement of the cylinders caused customer would have already implemented DCP and VVL based on dissatisfaction, and the technology was soon taken out of the cost/benefit ratio. This means that there is less pumping production. Since then, engine control systems and program- loss left to reduce, resulting in an incremental 1 to 2.5 percent ming ability have diminished the drivability concerns with reduction for a V6 and a 1.5 to 4 percent reduction for V8 modern day deactivation systems. New solutions have been configurations. The lower cost-benefit ratio for cylinder de- developed to address the NVH concerns that arise when activation makes the technology far less attractive on DOHC cylinders become deactivated. The NVH is a concern during engines. Despite the existence of prototype four-cylinder deactivation due to the “lower frequency, higher amplitude engines with cylinder deactivation, the committee believes torque pulsations at the crankshaft” (Leone and Pozar, 2001). the cost and customer dissatisfaction issues related to NVH With the addition of active engine mounts, any vibrations outweigh the benefits of implementing this technology on which would normally transfer to the passenger compart- four-cylinder engines. ment of the vehicle, causing customer dissatisfaction, are Vehicle OEMs estimate the cost for deactivation is ap- nearly eliminated. However, active engine mounts add cost. proximately $115. Vehicle integration items that mitigate Today’s trend toward overhead cam (OHC) valve trains has NVH issues may incur additional costs in the $140 range. an added a level of cost and complexity to integrate cylinder The cost of applying cylinder deactivation to OHC engines deactivation. is much higher, i.e., $340 to $400 because more complex and costly valve train elements must be changed. The EPA estimates the incremental RPE cost to be $203 (or $135 Implementation of Cylinder Deactivation cost) for six cylinders and $229 (or $153 cost) for eight The integration of a cylinder deactivation system varies cylinders (EPA, 2008) (both assuming an RPE multiplier depending on the engine layout. For overhead valve V8 of 1.5). NESCCAF quotes a literature RPE of $112 to $746 and V6 engines, this can be accomplished fairly simply by (NESCCAF, 2004) (or $75 to $497 cost). Martec estimates a modifications to the passages that supply oil to the valve manufacturing cost increase of $220 for a V6 DOHC engine lifters along with different valve lifters (Falkowski et al., (Martec Group, Inc., 2008). Sierra Research estimates an 2006). Implementation of a deactivation system on an OHC incremental cost of $360 to $440 (Sierra Research, 2008). engine is slightly different than on an OHV engine. One of EEA (2007) estimates for six-cylinder engines an RPE of the methods utilized for cylinder deactivation in an OHC $162 to $178 (or cost of $108 to $119) with an additional cost roller finger follower system involves the use of a switch- of $140 for NVH. For eight-cylinder engines, EEA estimates able roller finger follower. In the follower’s normal mode, an RPE of $205 to $225 (EEA, 2007) (or cost of $137 to the valve will operate as usual and maximum lift will still be $150 assuming an RPE of 1.5). The committee estimates that achieved. To deactivate the cylinder, a locking mechanism the manufacturing cost of implementing cylinder deactiva- must be released on the follower by oil pressure (Rebbert et tion for OHV would be $220 to $255 and $340 to $420 for al., 2008), collapsing the follower and rendering the valve engines with SOHC (not including RPE). inactive. Camless Valve Trains Fuel Consumption Benefit and Cost of Cylinder Deactivation A fully camless valve train eliminates the need for cam- Vehicle OEMs estimate cylinder deactivation typically shafts, as well as various other supporting hardware, and yields fuel consumption reductions in the 6 to 10 percent operates the valves individually by means of actuators. This range on V8 configurations. Testing done by FEV on a would allow for VEM fuel consumption saving technologies, V8 engine found that a decrease in fuel consumption of such as cylinder deactivation and continuously variable valve 7 percent occurred on the New European Drive Cycle lift and timing, to be applied all in one package. However, (NEDC). According to FEV, these reductions would be the complexity of the controls required makes for a diffi- “even higher for the US driving cycle, because of the US cult integration. Camless valve trains are electromagnetic, cycle’s higher proportion of part load operating conditions” hydraulic, pneumatic, or combinations of these that all face (Rebbert et al., 2008). NESCCAF estimates a 4 to 6 percent fundamental obstacles. By replacing the valve train, BMW reduction in fuel consumption (NESCCAF, 2004). The EPA claims the frictional saving from just the roller-bearing estimates a 6 percent reduction in fuel consumption (EPA, valve train achieves a further 2 percent reduction in fuel 2008). Sierra Research’s simulation estimated a reduction consumption. BMW also claims an overall reduction of up in consumption of 7.5 to 8.8 percent (Sierra Research, to 10 percent from camless operation (Hofmann et al., 2000). 2008). EEA estimates a 5.3 to 7.1 percent reduction in fuel However, none of these has been shown to offer advantages consumption (EEA, 2007). For OHV engines, the commit- not observed with the aforementioned cam-based systems. tee estimates a 4 to 6 percent drive-cycle fuel consumption The very high valve-timing precision associated with most reduction on a V6, and a 5 to 10 percent reduction on a V8. cam-driven systems is subject to compromise with camless For OHC engines, the committee assumes manufacturers approaches. The ballistic character of the valve assembly

OCR for page 38
48 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES with any camless system presents many control challenges. pumping losses, and lean overall mixture ratios to achieve In addition, the power demand for camless systems is gener- more thermodynamically efficient expansion processes. ally higher than that of their cam-driven counterparts. However, the TCCS and PROCO systems suffered from injec- Camless systems are perceived to have significant durabil- tor fouling, high exhaust emissions and low power density. ity risk, and as a result, no production implementations of Nonetheless, the goals of these engine systems remained camless systems have been announced. It is the judgment valid and interest returned to DISI following progress in of the committee that camless systems need further develop- fuel-injection systems and engine controls during the 1980s ment and are not expected on the market before 2015. and early 1990s. Mitsubishi introduced the first production implementation of DISI (which they called GDI) in Europe in 1996 (Iwamoto et al., 1997) in a 1.8-L four-cylinder engine, GASOLINE DIRECT INJECTION followed shortly after by a 3.5-L V6 in 1997. These GDI The most recent development of direct injection spark systems utilized lean-overall stratified-charge combustion but ignition (DISI) (also known as GDI) systems (Wurms et al., with some inlet throttling. It was soon found that typical in- 2002) have focused on early-injection, homogeneous-charge use fuel consumption was significantly higher than European implementations using stoichiometric mixture ratios under emissions-test-schedule results suggested. most operating conditions. These conditions allow for the Following an initial burst of interest, Mitsubishi GDI sales use of highly effective and well-proven closed-loop fuel con- were lower than expected. Hence, this system was withdrawn trol and three-way catalyst exhaust aftertreatment systems. from the market, and there was a return to conventional PFI Fuel consumption benefits of these homogeneous versions systems. It was believed that this withdrawal stemmed not are derived mainly from a knock-limited compression ratio only from disappointing sales but also because meeting up- increase (typically +1.0) enabled by forcing all of the fuel coming NOx emissions standards in Europe and especially to vaporize in the cylinder. This yields a charge-cooling ef- the United States using only combustion system control was fect that suppresses the knocking tendency. Another added more difficult than anticipated, and lean NOx aftertreatment benefit of charge-cooling is an increase in the volumetric systems were seen as very costly and of questionable reli- efficiency from the increase in density of the incoming ability for volume production. charge. In contrast, with port fuel injection (PFI) systems some of the fuel vaporizes in the intake port, and this conveys Implementation of Direct Injection heat from outside of the cylinder, i.e., from the intake port, to the in-cylinder charge. While heating of the intake charge A concern today (as in the past) with DISI systems is the is a negative (relative to the knock-limited compression ratio matter of fuel-based carbonaceous deposits forming from and performance) it does provide a measure of “thermal residual fuel in the injector nozzle upon hot engine shutdown. throttling” at typical road loads, which reduces negative Carbonaceous deposits can restrict fuel flow and also modify pumping work. Thermal throttling, like common pressure fuel-spray geometry in some unfavorable manner (Lindgren throttling, lowers the mass of inducted fuel-air mixture thus et al., 2003). Locating the injector in a relatively cool part of reducing power, which is the objective of throttling. It does the cylinder head is one approach to alleviating this problem. this, however, with less pumping loss than the conventional Fuel variability in the United States is of some concern rela- throttling used with homogeneous DISI. tive to this issue based largely upon the olefin content of the In terms of additional losses, DISI relies on fuel pressures fuel, which typically is higher than that found in European that are higher than those typically used with PFI systems gasoline. While some concerns with deposits remain, they (e.g., 150-200 bar versus 3-5 bar for PFI), and the increase are being alleviated mainly by injector design improvements. in required fuel pump work increases parasitic loss. Finally, DISI researchers often make reference to wall-guided, these homogeneous, stoichiometric DISI systems cannot ex- flow-guided, or spray-guided injection (Kuwahara et al., ploit the thermodynamic expansion efficiency gains possible 2000), and in general these terms refer to different geometric with lean overall mixtures. arrangements of the fuel injection and mixture preparation processes. For example, wall-guided usually refers to place- ment of the fuel injector to the side of the cylinder near the History of Direct Injection corner of the cylinder head with the cylinder wall. The spray Early (1960s and 1970s) versions focused on late-injection, is then aimed across the cylinder toward the top of the piston lean overall stratified-charge implementations as exemplified when the piston is near the top of the cylinder. In this case by the Texaco TCCS (Alperstein et al., 1974) and Ford the piston crown shape is the “wall” which guides the spray PROCO (Simko et al., 1972) systems, neither of which (Kuwahara et al., 2000). In spray-guided engines, the injector entered volume production. These systems were attempts to is located in the cylinder head near the center of the cylinder utilize gasoline and other fuels in spark-ignited engines de- with the spray aimed down the cylinder axis (Schwarz et al., signed to take advantage of two of the three thermodynamic 2006). Injection in this case would be timed later during the advantages of diesels, namely lack of throttling to eliminate induction process. The fuel-spray trajectory is then guided

OCR for page 38
50 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES because it occurs during valve overlap. This synergism of those with direct fuel injection) it has been found that cooled turbocharging, DISI, and blow-through can enable further EGR can be seen as an alternative means for controlling engine downsizing, and an additional fuel consumption ben- knock at moderate engine speeds and medium to high loads. efit may thus result. Unfortunately, this engine performance Under certain operating and base-engine conditions, passing opportunity occurs in the knock-sensitive operating range. As the EGR through a heat exchanger to reduce its temperature a result, establishing acceptable vehicle launch performance can be a more fuel-efficient means of controlling knock with turbocharged and downsized engines is challenging. compared to spark-timing retardation and fuel-air ratio en- The distinction between research octane number (RON) richment. The fuel consumption benefits of this feature are and motor octane number (MON) is particularly noteworthy highly dependent upon the base engine to which it is applied when fuels other than traditional gasoline are considered. and the engine’s operating map in a particular vehicle. As The test methodology on which RON is based reflects resis- the heat exchanger must be equipped with a diverter valve tance to thermal auto-ignition resulting from both chemical to accommodate heat-exchanger bypass for lighter-load and heat-of-vaporization (evaporative cooling) properties, operation, the sequences of carbonaceous deposit formation whereas MON is relatively insensitive to the latter of these. in the heat exchanger, in the diverter and control valves, The difference between these two metrics is termed sensi- and in the turbine are among the real-world factors that tivity (RON – MON = sensitivity). When fuels like ethanol can compromise the overall performance of this feature. are considered, the aforementioned distinction should be This feature is in production for CI engines for which the emphasized as this fuel has a very high RON, but its MON exhaust particulate level is much higher than for downsized is moderate. Hence, the sensitivity of ethanol is 18, whereas and boosted SI engines; however, packaging the system into that of a typical gasoline is considerably lower, e.g., 10. The certain vehicles can make implementation difficult. consequence of high-sensitivity fuels when aggressive boost- Variable geometry turbochargers (VGTs), commonly ing and high compression ratios are pursued is an increased used on CI diesel engines, have not reached mainstream use vulnerability to pre-ignition problems. This typically results on SI engines. The concern with using VGTs on gasoline- from engine operation in the peak-power range where all engine exhaust has been the ability of the adjustable blades surface temperatures to which the fuel is exposed are very and their adjustment mechanism to withstand the higher high. This tends to reduce the heat-of-vaporization benefit as- temperatures of the gasoline exhaust gases. A diesel engine sociated with ethanol. It has been widely recognized for most typically has lower exhaust gas temperatures, and material of the history of the SI engine that water induction along selection for the adjustable blades has been successful in pro- with the fuel and air can reduce the thermal auto-ignition duction. Recently, Porsche and Borg Warner have developed tendency and thus can increase the torque and power output. a variable geometry turbo to be used on the Porsche 911. While this has been widely used in racing communities, there This turbocharger required the development of new material are some practical limitations to the general applicability specifications that could withstand the higher temperatures of of this, e.g., water can find its way into the crankcase and the exhaust gases. Due to the high cost of material to with- form an emulsion with the oil and therefore compromise the stand the heat and ensure long-term functionality of the vane lubrication system. guides, VGTs are currently seen only for use in high-end The evaporative characteristic of any liquid largely de- vehicles. Alternatively, a downsized, fixed-geometry turbo- pends upon intermolecular affinity, and in the cases cited charger may be used, but this approach will compromise above the so-called hydrogen bonding is a major component. power output because the fixed exhaust turbine geometry This involves the polarized bonds between hydrogen and will restrict airflow through the engine in order to provide oxygen atoms where there is a slight positive charge on the acceptable low-speed turbocharger transient response. Extra- hydrogen atom that is bound to an adjacent oxygen atom, slippery torque converters (e.g., those with higher stall speed) which carries a slight negative charge. Hence, the positive can help to alleviate turbo lag issues, but they will also charge on the hydrogen atom of the −OH group applies impose a fuel consumption penalty from increased slippage. an attractive force acting on the negative charge on the General Motors performed simulation testing on its 2.4-L oxygen atom of a nearby molecule. This grouping of −OH- port fuel-injected four-cylinder engine in the Chevrolet containing molecules, be they ethanol or water, is responsible Equinox. The port fuel-injected 2.4-L engine was compared for their relatively high evaporative-cooling characteristic. to an engine of the same displacement equipped with direct This evaporative cooling characteristic can be utilized to injection, turbocharger, and dual VVT. GM claims that this prevent knock at certain engine operating conditions by approach “can improve fuel consumption on the FTP cycle implementing a system that can selectively inject the charge by up to 10 percent relative to an engine with VVT” but cooling liquid. This system is discussed below in this chapter without DI and turbocharging (EEA, 2007). in the section “Ethanol Direct Injection.” Ford Motor Company has been developing downsized Exhaust-gas recirculation (EGR) is well known as a and turbocharged engines equipped with direct injection. means to reduce pumping losses and thereby increase fuel The company plans to offer these engines in nearly all its efficiency. With downsized turbocharged engines (including upcoming models in the future. One of the engines is 3.5 L

OCR for page 38
51 SPARK-IGNITION GASOLINE ENGINES for gasoline VGTs. System detail choices depend largely on in displacement and features twin turbochargers with direct vehicle performance targets. Martec estimates that the manu- injection. From testing, Ford has claimed that this engine will facturing cost of downsizing a six-cylinder to a turbocharged reduce fuel consumption by 13 percent when compared to a four-cylinder engine is $570, and a downsize from an eight- V8 with similar performance (EEA, 2007). cylinder to a six-cylinder turbo adds a manufacturer cost of $859 (Martec Group, Inc., 2008). For the six-cylinder to a Fuel Consumption Benefit and Cost of Downsizing and four-cylinder case, Martec is including a $310 downsizing Turbocharging credit and a $270 credit for eight cylinders to six cylinders. Martec’s system price includes a water-cooled charge air The EPA estimates that a fuel consumption reduction of cooler, split scroll turbo, and upgraded engine internals 5 to 7 percent can occur with downsizing and turbocharg- (not including “modifications to cylinder heads, con-rods, ing (EPA, 2008). This estimate assumes that the vehicle is and piston geometry or coatings”) (Martec Group, Inc., currently equipped with a DISI fuel system. NESCCAF 2008). It should be noted that most manufacturers tend to estimates a 6 to 8 percent reduction in fuel consumption use air-cooled charge air coolers. Sierra research estimates (NESCCAF, 2004). A study performed by Honeywell Turbo an incremental RPE adjusted cost increase of $380 to $996 Technologies estimates that a 20 percent reduction in fuel (Note: values have been adjusted from Sierra’s 1.61 RPE consumption is possible from downsizing by 40 percent factor to 1.5) (Sierra Research, 2008). Sierra’s price estimate (Shahed and Bauer, 2009). FEV claims by downsizing and is based on a “relatively simple turbocharger system that turbocharging a consumption reduction of 15 percent can would not be able to match the launch performance of the occur in the New European Drive Cycle. An additional 5 to larger, naturally aspirated engine.” The value provided by 6 percent is possible with the addition of a DI fuel system Sierra is “not including the catalyst plus $650 in additional (Ademes et al., 2005). The expected consumption reductions variable cost for a turbo system marked up to RPE using a are highly load dependent. The highest benefits will occur factor of 1.61” (Sierra Research, 2008). The EPA provided at low load conditions. Reduction in consumption is due to incremental costs for large cars, minivans, and small trucks higher engine loads and lower friction loss. Sierra Research at $120. This cost included a downsizing credit. For the small estimates midsize sedans will increase fuel consumption by car classification, the EPA has estimated an incremental cost 0.3 percent and pickup trucks will decrease consumption by of $690. The higher cost for the small car is due to the lack 0.3 percent (Sierra Research, 2008). Sierra’s values are lower of significant engine downsizing possibilities (EPA, 2008). than others since Sierra did not increase the octane require- EEA estimates a V6 approximately 3 liters in displacement ment for the engine or combine it with direct injection. Sierra to have an RPE adjusted cost of $540 (or $360 cost assum- was therefore forced to lower the compression ratio in order ing an RPE factor of 1.5) (EEA, 2007). Pricing for the EEA to reduce the knocking tendencies while avoiding an octane study was based on a standard turbo, air-to-air intercooler, requirement increase. Sierra claims that “turbocharging and engine upgrades, additional sensors and controls, and intake downsizing without the use of gasoline direct injection does and exhaust modifications. not yield benefits on a constant performance basis, based on The committee estimated that the manufacturing costs a statistical analysis of available CAFE data done in 2004” for integrating downsizing and turbocharging would be (Sierra Research, 2008). The committee concluded that for in the range of a $144 cost savings to a $790 additional the purposes of this report, turbocharging and downsizing will cost, depending on the engine size and configuration. See always be applied following DI in order to minimize the need Table 4.A.1 in the annex at the end of the chapter for a to reduce compression ratio. This order of implementation complete breakdown of cost benefits for each engine size. is in agreement with recent industry trends. The committee The teardown studies currently being performed for the EPA estimates that a 2 to 6 percent reduction in fuel consumption by FEV (Kolwich, 2009, 2010) have been deemed the most is possible when downsizing and turbocharging is added to accurate source of cost information by the committee, and an engine with DI. therefore these studies were the primary source used for There is a large variation in the cost estimates from the these cost estimates. As with other sources, the committee various sources, which arises from a couple of key items. encourages the reader to view the original document to gain One item is whether or not there is a credit included in the a better understanding of how the costs were derived. The cost from decreasing the engine cylinder count (e.g., going cost increase for an I4 is somewhat obvious, due to the cost of from V6 to I4) and the amount of the credit. Another source additional components and a lack of significant downsizing of difference is from the use of a split scroll turbine housing credit. The downsizing credit is small because the cylinder or a standard housing on the turbocharger. The split scroll count remains the same and generally the same number of adds cost compared to the standard-type housing. valve train, fuel system, and other supporting components Vehicle OEM input indicates that basic, fixed-geometry are still required. The very low cost of converting from a turbochargers add roughly $500 system cost, and dual-scroll DOHC V6 to a turbocharged DOHC I4 is due to the very turbocharger systems can add about $1,000 (not considering large downsizing credit from removing two cylinders and an RPE factor). Currently no pricing information is available

OCR for page 38
52 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES Piston-Assembly Friction the supporting hardware for a whole bank of the engine, such as moving from four camshafts to two. In this report, Piston-assembly friction is a major component of overall the conversion from a Vee-type engine to an in-line is used engine friction, and of this the oil-control ring is the biggest only when moving from a V6 to an I4, as an I6 (from a V8) contributor. Efforts have been underway for several decades is far less common in the market. When converting from to minimize the radial dimension of the rails to render them a V8 to a V6, the downsizing credit is much smaller, as more conformable, with minimum spring force, to bores you lose two cylinders but still have a Vee engine with two that may not be perfectly circular. Unlike oil-control rings, banks requiring two cam drive systems, four camshafts, etc. which are forced against the cylinder liner surface only by Also, turbocharging a V6 usually requires a more expensive their expander spring, the forces pushing the compression twin-turbo system, versus the single turbo on the I4. To rings against the cylinder are gas-pressure forces in the ring summarize, the downsizing credit is much smaller and the groove behind the rings. This gas pressure comes from the turbocharging cost is much higher for going from a V8 to a cylinder gases that pass down into the ring groove by way of V6 than for going from a V6 to an I4. the ring end gap, and little can be done to reduce the frictional contribution of compression rings. It should be noted that it is ENGINE FRICTION REDUCTION EFFORTS only during the high-pressure portions of the cycle that their frictional contribution is significant. It is noteworthy that Engine friction can account for up to 10 percent of the bore distortion either due to thermal distortion of the cylinder fuel consumption in an IC-powered vehicle (Fenske et al., block when the engine heats up to operating temperature or 2009). Therefore, reducing friction is a constant aim of to mechanical distortion caused by the forces resulting from engine development for improved fuel economy. A large torquing the cylinder-head attachment bolts must be mini- majority of the friction in an IC engine is experienced by mized if ring friction is to be minimized (Abe and Suzuki, three components: piston-assembly, bearings (i.e., crankshaft 1995; Rosenberg, 1982). journal bearings), and the valve train. Within these compo- nents friction comes in two general forms: hydrodynamic Crankshaft Offset viscous shear of the lubricant (mainly in journal bearings) and surface contact interactions, depending on the operating Crankshaft offset from the cylinder centerlines will alter conditions and the component. connecting-rod angularity. If this is done in a manner that There are several approaches to reduce frictional losses reduces the piston side loading during the high-pressure por- in an SI engine, mainly through the design of the engine tion of the engine cycle (i.e., the expansion stroke), a piston- and lubricant. A common trend has been to utilize low- skirt friction reduction is theoretically possible. Some early viscosity lubricants (LVL) to reduce energy loss through 20th-century engines employed this concept, and some rela- lowered viscous shear (Nakada, 1994); significant fuel tively recent claims have been made on this design strategy. economy improvements have been demonstrated through Recent efforts to document any friction reduction have failed this adaptation (Taylor and Coy, 1999; Fontaras et al., 2009). to show any benefit (Shin et al., 2004). It is likely that the However, lowering viscosity also effectively reduces the tribological state at the piston-skirt-to-cylinder-wall interface lubricant thickness between interacting component sur- will affect this, i.e., presence or absence of a hydrodynamic faces, which can increase the occurrence of surface contact. oil film in the critical area under typical operating conditions. Increased surface contact can have the detrimental effect of increased wear and heat generation, which can in turn Valve Train Friction affect engine durability. In addition to lowered lubricant vis- cosity, other SI technology trends (in particular turbo charg - Valve train friction underwent a major reduction in ing and downsizing) lead to increased power density, which the mid-1980s with near-universal adoption of roller cam can cause increased surface interaction (Priest and Taylor, followers. Valve-spring tension reduction may also reduce 2000). In order to maintain engine durability, improving valve train friction, but reduction down to the valve-motion mixed lubrication performance in vulnerable components dynamic-stability limit have been found to yield susceptibil- should be considered. Improvements in lubricant additives ity to compression loss under circumstances where carbona- (low friction modifiers) and surface engineering (surface ceous deposits become detached from chamber surfaces and coatings and surface topography design) are methods that become trapped between the valve seat and valve face have been used to improve performance in these surface and thus cause major valve leakage. contact conditions (Erdemir, 2005; Etsion, 2005; Sorab et al., 1996; Priest and Taylor, 2000). Crankshaft Journal Bearing Friction The following sections discuss in more detail specific engine design considerations for reducing friction, and also Energy loss due to crankshaft journal bearing friction provide further discussion of low-viscosity lubricants. tends to scale as the cube of the diameter times the length, or

OCR for page 38
53 SPARK-IGNITION GASOLINE ENGINES (diameter)3 × (length). Efforts are always made to minimize tional friction reduction can be achieved through engine this source of friction, but adequate crankshaft stiffness at the component design and through improvements of surface en- pin-to-main joints and overall length constrain this option. In gineering (surface coatings, material substitutions, selective V6 engines adequate pin-to-pin joint strength integrity must surface hardening and surface topography control). The EPA also be maintained. estimated potential FC benefit at a range of 1 to 3 percent with a cost of $7 per cylinder (EPA, 2008). Given recent advancements in engine friction reduction, the committee Low-Viscosity Lubricants estimates that the potential FC benefit is 0.5 to 2.0 percent As discussed previously, lowering lubricant viscos - at a manufacturing cost of $8 to $13 per cylinder. ity reduces viscous shear. Therefore moving to advanced low-viscosity lubricants has the potential to improve fuel ENGINE HEAT MANAGEMENT economy; however, there is debate about the range of ef- fectiveness. Several studies have examined the effectiveness As there is never a shortage of waste heat in and around IC of LVL in lowering friction and reducing fuel consumption engines, efforts to utilize this in productive ways have been (Sorab et al., 1996; Taylor and Coy, 1999; Fontaras et al., ongoing for decades. Following are some methods of im- 2009). Variations in test methodologies, i.e., vehicle fuel proving heat management; however, these techniques are not consumption measurement versus engine-dynamometer assigned a fuel consumption benefit or cost for this analysis. motoring tests, have led to some confusion in this area. Sorab tested the effectiveness of low-viscosity lubricants on Piston-Crown Design one component of an IC engine, the connecting rod journal bearing. Experimental testing showed significant friction Piston-crown design can affect its temperature. In some reduction; however, it is difficult to extend these results to cases moving the piston-ring pack upward motivated by an overall fuel consumption benefit. Taylor and Coy (1999) hydrocarbon-emissions reduction efforts to reduce crevice reviewed several modeling techniques that analyzed the fuel volume also tended to reduce piston-crown temperatures consumption benefit of designed lubricants. It was shown and thus reduced the knock tendency in some cases. To the that lubricants with designed low-viscosity properties can extent that this enabled a small increase in compression ratio, reduce FC by up to 1 percent. Fontaras et al. (2009) tested a small fuel consumption benefit may result along with a the fuel consumption benefit of LVL in different drive cycles. significant reduction in hydrocarbon emissions. In some The benefit ranged from 3.6 percent down to negligible cases this piston modification shortened the heat-conduction depending on the driving cycle. For a cycle that includes pathway by which heat in the piston crown is transferred a cold start, the LVL effectiveness is higher since the low- through the second piston land and then into the top ring and temperature viscous behavior prevails in this cycle. In a fully to the cylinder and into the coolant. warmed-up engine the FC benefits are not as noticeable and can even be negligible. Cylinder-Temperature Profile Cylinder-temperature profile has been found to have Fuel Consumption Benefit and Cost of Reducing Engine subtle effects on efficiency. If the upper portion of the Friction cylinder can be made to run cooler and the lower portion The effectiveness of low-viscosity lubricants has limited hotter, then both friction and hydrocarbon emissions may drive cycle testing. Fontaras et al. (2009) performed several benefit. This result can readily be achieved by shortening the tests of LVL over different drive cycles, with the conclusion coolant jacket such that only about 75 percent of the piston that a benefit of 1 to 1.5 percent can be achieved without stroke equivalent is cooled by the coolant. At a fixed coolant affecting the overall engine performance. It was noted that pump capacity, higher coolant flow velocities are available the actual consumption reduction will vary by the amount at the top of the cylinder. This can enable an overall friction of time spent in transient operation and if the drive cycle is reduction by reducing the extent of boundary-layer piston one in which the engine must be started cold (Fontaras et al., ring friction at the top and a lubricant viscosity reduction at 2009). The EPA estimated that a reduction in consumption the bottom of the stroke. In addition, the higher temperature of 0.5 percent can occur with the use of LVL at a cost of of the lower portion of the cylinder promotes post-flame oxi- $3 per vehicle (EPA, 2008). Considering the more relevant dation of the fuel-air mixture that leaves the piston top-land U.S. drive cycle and the current widespread use of 5W30, crevice late in the expansion stroke. the committee estimates that an additional 0.5 percent FC benefit can be realized with more advanced synthetic LVL Exhaust Port Surface Area at a cost of $3 to $5 per vehicle. Improved engine friction reduction is a constant aim, yet Exhaust port surface area can affect the heat input to there is still opportunity for additional FC benefit. Addi- the cooling system, and this has subtle efficiency and ex -

OCR for page 38
54 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES haust emissions consequences. A significant portion (~50 were shown to the public (Alt et al., 2008) suggesting that percent) of the heat that enters the cooling system does so controls-related progress has been made. As system defini- by way of the exhaust port. Typically, the high temperature tion, fuel consumption benefits, and costs are uncertain at of the exhaust that leaves the cylinder at the beginning of this time, HCCI is believed to be beyond the 15-year time the exhaust-valve open period is also characterized by its horizon of this study. highly turbulent state. The associated high rates of heat transfer can affect both the heat load on the cooling sys - COMBUSTION RESTART tem as well as the time required for the catalyst system to achieve operating temperatures following cold start. It is Combustion restart can be seen as an enabler for idle-off noteworthy that at peak power the highest exhaust flows operation, which has the potential to reduce fuel consump- occur during the blowdown process when the valve flow tion under drive conditions that have significant idle time. area is a limiting factor, and when the valve is fully open The principle challenge relates to the crankshaft position near mid-exhaust stroke, the so-called displacement flow when the engine comes to rest. One cylinder must be in the is somewhat lower. early phase of the expansion stroke such that fuel can be in- Typically if the exhaust-port cross-sectional area is re- jected via DISI and spark(s) delivered to initiate combustion duced until there is evidence of incremental exhaust pumping and expansion with sufficient potency to initiate sustained work under peak power operating conditions, no power loss engine rotation. Overcoming the aforementioned challenge is to be expected. Efforts to reduce exhaust-port surface area is highly dependent upon many real-world conditions over may reduce the heat load on the cooling and also cause the which there are limited opportunities without the addition of exhaust temperatures to be somewhat higher. This can yield some form of electro-machine to properly position the crank- a fuel consumption benefit if ignition-timing retardation, shaft prior to restart. Given this challenge, it is believed that which is often used to facilitate rapid catalyst light-off, can this approach will not attain significant market penetration be minimized. A downsized coolant pump, cooling fan, and during the time horizon of this study. radiator core may also be beneficial. ETHANOL DIRECT INJECTION Electrically Driven Coolant Pumps An approach to cooling the charge to control knock and Electrically driven coolant pumps are also frequently detonation ties in with both the octane ratings of fuels mentioned as fuel consumption enablers. While these tend and their heats of vaporization. This approach is to inject to decrease parasitic loads during warm-up, local hot spots into the intake charge or into the cylinder a fluid with a larger may cause bore and valve-seat distortion or gasket failures. heat of vaporization than the fuel itself. This fluid would then Fuel consumption reduction derived from the above items vaporize drawing the heat of vaporization from the intake depends on the details of the initial engine design. A more or cylinder gases thus lowering their temperature. Direct- detailed discussion of the electrification of water pumps can injected (DI) E85 (i.e., a mixture of ~85 percent ethanol be found in Chapter 5 of this report. and ~15 percent gasoline) has recently been proposed for use both as an anti-knock additive and as a way to reduce petroleum consumption (Cohn et al., 2005) for boosted SI HOMOGENEOUS-CHARGE COMPRESSION IGNITION engines. A recent in-depth study of this concept was carried While homogeneous-charge compression ignition (HCCI) out at Ford (Stein et al., 2009) where engine dynamometer has received much attention in the recent past, some funda- studies were carried out with a turbocharged 3.5-L V6 engine mental control-related challenges remain. The absence of a using gasoline PFI combined with DI E85. The promise of discrete triggering event in close temporal proximity to the this approach is to enable three benefits, namely, allowing desired time of combustion is the basis for these challenges. increasing the compression ratio of the boosted engine; In this type of combustion system, temperature is all impor- allowing increasing the level of boost usable without knock tant; many real-world factors can come into play that will and pre-ignition limitations; and enabling operation closer to yield unexpected outcomes, e.g., previous-cycle effects and MBT, timing. These three benefits provide greater thermal piston and valve temperature swings. As HCCI combustion efficiency as well as increased power, which allows further is essentially instantaneous, it produces very high rates of downsizing and downspeeding, thus adding potential fuel pressure rise and high peak pressures. Engine structural at- consumption reductions. The Stein et al. study (2009) used tributes must take this into account. a prototype V6 DI turbocharged engine (termed Ecoboost Unthrottled HCCI combustion at light loads may produce by Ford) with a PFI gasoline injection system added to the very high hydrocarbon emissions when the exhaust-gas original direct-injection fuel system. The DI fuel system was temperature is relatively low, and this may challenge exhaust separated from the PFI system and supplied only with E85 aftertreatment processes. Nonetheless, advanced prototype from a separate tank and pump. The engine was operated vehicles using HCCI over a portion of the operating range at both the base 9.8:1 compression ratio and a high value

OCR for page 38
55 SPARK-IGNITION GASOLINE ENGINES FINDINGS of 12:1. E85 injection quantities and spark advance were optimized, and measured results were then extrapolated to SI engines are widely accepted as the primary source of application with a 5.0-L engine in a pickup truck by means propulsion for light-duty vehicles in the United States. There of full system simulation. The anticipated benefits were have been significant improvements in the fuel consumption observed. Namely, MBT spark timing was achievable up to reduction of SI engines in response to past trends of rising higher loads than were possible without the E85 injection, fuel prices. These improvements are in large part due to past leading to a reduction in both gasoline and overall (combined advancements in fast-burn combustion systems with strategic gasoline and E85) fuel consumption. One of the conclusions exhaust-gas recirculation (EGR), multi-point fuel injection, reached by Stein et al. (2009) was the following: and reduced engine friction. Newly available SI technologies are assessed with respect to fuel consumption benefit and By enabling increased CR [compression ratio], engine down- cost measured against the aforementioned technologies as sizing, and downspeeding, E85 DI + gasoline PFI makes the the baseline. These current technologies address improve- engine more efficient in its use of gasoline, thereby leverag- ments in the areas of continuing friction reduction, reduced ing the constrained supply of ethanol in an optimal manner pumping losses through advanced VEM, thermal efficiency to reduce petroleum consumption and CO2 emissions. For a improvements, and improved overall engine architecture, hypothetical 5.0 L E85 DI + gasoline PFI engine in a Ford F-series pickup, the leveraging due to 12:1 CR is approxi- including downsizing using turbocharging and GDI. The mately 5:1 on the EPA M/H drive cycle. That is, 5 gallons significant finds are as follows: of gasoline are replaced by 1 gallon of E85. This leverag- ing effect will be significantly reduced for more aggressive Finding 4.1: SI technologies offer a means of reducing fuel drive cycles. consumption in relatively small, incremental steps. OEMs can thus create packages of technologies that can be tailored to Since the focus of the present report is reducing petroleum meet specific cost and effectiveness targets. It is the combina- consumption, the implications of the Stein et al. work on op- tion of numerous, affordable SI technologies in a package that timizing ethanol utilization will not be considered. However, makes them appealing when compared to diesel or full hybrid the combination of increased compression ratio as well as alternatives—which offer a single large benefit at a large cost. downsizing and increased boosting possible with the ethanol Because of this capability, and considering the wide accep- injection enables reducing fuel consumption compared with tance of SI engine applications, the committee believes that operation on gasoline alone. the implementation of SI engine technologies will continue to Any approach to inject an anti-knock fluid such as E85 play a large role in achieving reduced levels of fuel consump- would require an additional tank on the vehicle to provide the tion. Table 4.A.1 at the end of this chapter summarizes the anti-knock fluid for injection and would require a willingness fuel consumption reductions and costs for these technologies. on the part of the vehicle driver to fill the anti-knock fluid tank. In the study by Stein et al. (2009), the authors estimated Finding 4.2: Cylinder deactivation is most cost-effective based on vehicle simulations for a full-size pickup truck that when applied to OHV V6 and V8 engines; it typically affords E85 usage on the FTP urban/highway schedule would be 4 to 10 percent fuel consumption reduction. The higher cost of only about 1 percent of the total fuel used, thus providing an applying cylinder deactivation to DOHC V6 and V8 engines, E85 refill driving range of ~20,000 miles with a 26-gallon combined with the reduced fuel consumption benefit when gasoline fuel tank and a 10-gallon E85 tank. For the higher- cylinder deactivation is added to an engine with VVT, has load US06 driving cycle, E85 would constitute 16 percent caused most OEMs to avoid its application to DOHC engines. of the fuel used for an E85 refill range of ~900 miles. For For this reason, the committee believes that cylinder deacti- towing a trailer up the Davis Dam slope (~6 percent grade vation will be applied only to OHV engines in most cases. for over 10 miles), E85 usage would be 48 percent of the fuel used with an E85 tank refill range of ~100 miles. Once Finding 4.3: Stoichiometric gasoline direct injection (SGDI) all the anti-knock fluid has been consumed, spark timing applied to naturally aspirated engines typically affords a would have to be retarded and turbocharger boost reduced to knock-limited compression ratio increase of 1.0 to 1.5 and a prevent knock if a high compression ratio were chosen for the reduction in fuel consumption of 1.5 to 3.0 percent at a cost engine (e.g., 12 versus 9.8) based on reliance on injection of of $117 to $351, depending on cylinder count and including an anti-knock fluid to control knock. Operating with retarded noise-abatement items. Versions of direct injection that pro- spark timing and reduced boost would not harm the engine vide some measure of charge stratification can further reduce but may impact available power. fuel consumption, but emissions and implementation issues Based on the costs for the urea dosing systems used for have inhibited high-volume applications. CI engine selective catalytic reduction aftertreatment that has similar componentry (see Chapter 5), the cost of converting Finding 4.4: Turbocharging and downsizing, while main- a boosted DI engine to PFI gasoline with DI E85 injection taining vehicle performance, can yield fuel consumption is estimated to be $300 to $350.

OCR for page 38
56 ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES BIBLIOGRAPHY reductions ranging from 2 to 6 percent, depending on many implementation details such as changes in cylinder count. Abe, S., and M. Suzuki. 1995. Analysis of cylinder bore-distortion during Industry trends and input from OEMs show that this tech- engine operation. SAE Paper 950541. SAE International, Warrendale, nology is usually added in combination with direct injection Pa. when the goal is improved efficiency. SGDI will help negate Ademes, N., O. Lang, S. Lauer, W. Salber, and H. Jene. 2005. Valve train variability for advanced gasoline engines. MTZ 66(12; December). the need to reduce compression ratio when turbocharging, Alperstein, M., G. Shafer, and F. Villforth. 1974. Texaco’s stratified-charge- giving the combination a positive synergistic effect. If the engine—Multifuel, efficient, clean, and practical. SAE Paper 740563. cylinder count is reduced, NVH-related issues will reduce SAE International, Warrendale, Pa. the benefit level. Cost estimates were based primarily on Alt, M., U.D. Grebe, J.R. Dulzo, V.A. Ramappan, P. Kafarnik, and P.M. Najt. the 2009 EPA teardown study and range from around zero 2008. HCCI—From lab to the road. 29th Vienna Motor Symposium. Automobile Magazine. 2002. Fiat multiair technology increases power, additional cost when converting from a V6 to an I4, to about l owers emissions. Available at http://rumors.automobilemag. $658 when converting from a V8 to a V6. com/6476771/green/fiat-multiair-technology-increases-power-lowers- emissions/index.html. Finding 4.5: The VEM over the speed-load range of an Bernard, L., A. Ferrari, R. Rinolfi, and C. Vafidis. 2002. Fuel economy SI engine can further reduce the pumping loss over that improvement potential of UNIAIR throttleless technology. SAE Paper 2002-33-0012. SAE International, Warrendale, Pa. of the previously described configurations and can also Boggs, D.L., H.S. Hilbert, and M.M. Schecter. 1995. The Otto-Atkinson- cause a slight increase in engine performance, which will Cycle engine: Fuel economy and emissions results and hardware design. offer a potential downsizing opportunity. There are many SAE Paper 950089. SAE International, Warrendale, Pa. different implementations of this, and the cost-benefit rela- Cleary, D., and G. Silvas. 2007. Unthrottled engine operation with variable tionship for these implementations depends on the engine intake valve lift, duration, and timing. SAE Paper 2007-01-1282. SAE International, Warrendale, Pa. architecture to which they are applied. Fuel consumption Cohn, D.R., L. Bromberg, and J.B. Heywood. 2005. Direct Injection Ethanol reduction can range from 1.0 percent with only intake cam Boosted Gasoline Engines: Biofuel Leveraging for Cost-Effective phasing, to about 11 percent with a continuously variable Reduction of Oil Dependence and CO2 Emissions. MIT Laboratory for valve lift and timing setup. The total cost range is $35 to the Environment Report LFEE 2005-01 RP. Massachusetts Institute of $530, depending on the implementation technique and en- Technology, Cambridge, Mass. Delphi. 2009. Gasoline engine management systems. Available at http:// gine architecture. delphi.com/manufacturers/auto/powertrain/gas/valvetrain/vcp/. EEA (Energy and Environmental Analysis, Inc.). 2007. Update for Ad - Finding 4.6: It is important to note that, according to in- vanced Technologies to Improve Fuel Economy of Light Duty Vehicles. dustry trends and input from OEMs, the major OEMs are Prepared for U.S. Department of Energy. August. Arlington, Va. either pursuing advanced VEM technologies, such as CVVL EPA (U.S. Environmental Protection Agency). 2008. EPA Staff Techni- cal Report: Cost and Effectiveness Estimates of Technologies Used (Nissan, Toyota, and Honda), or turbocharging and down- to Reduce Light-Duty Vehicle Carbon Dioxide Emissions. EPA420- sizing with SGDI (Ford, GM, and VW), but usually not both R-08-008. March. Washington, D.C. (aside from BMW, which has both on its new N55 engine). Erdemir, A. 2005. Review of engineered tribological interfaces for improved However, there would still be a benefit, diminished some- boundary lubrication. Tribology International 38:249. what by synergistic effects, to be gained by adding VVL to Etsion, I. 2005. State of the art in laser surface texturing. Journal of Tribology 127(January):248. a turbo/SGDI engine with VVT. The committee concluded Falkowski, A., M. McElwee, and M. Bonne. 2006. Design and develop- that thus far the industry has deemed the cost-benefit ratio ment of the DaimlerChrysler 5.7 L Hemi® engine multi-displacement too small to implement both technologies on one engine for cylinder-deactivation-system. SAE Paper 2004-01-2106. SAE Interna- Interna- mainstream vehicles. Adding continuously variable valve tional, Warrendale, Pa. lift and timing to a baseline DOHC engine with intake Fenske, G., R. Erck, O. Ajayi, A. Masoner, and A. Comfort. 2009. Impact of friction reduction technologies on fuel economy for ground vehicles. cam phasing can result in a 5 to 9 percent reduction in fuel Proceedings of the 2009 Ground Vehicle Systems Engineering and consumption. Implementing dual cam phasing, SGDI, and Technology Symposium (GVSETS). turbocharging and downsizing to a baseline DOHC engine Flierl, R., and M. Kluting. 2000. The third generation of valvetrains—New with intake cam phasing can provide a 6 to 11 percent fuel fully variable valvetrains for throttle-free load control. SAE Paper 2000- consumption reduction. 01-1227. SAE International, Warrendale, Pa. Flierl, R., D. Gollasch, A. Knecht, and W. Hannibal. 2006. Improvements to a four cylinder gasoline engine through the fully variable valve lift Finding 4.7: Variable compression ratio, camless valve and timing system univalve. SAE 2006-01-0223. SAE International, trains, homogeneous-charge compression ignition, and Warrendale, Pa. cooled EGR were all given careful consideration during the Fontaras, G., E. Vouitsis, and Z. Samaras. 2009. Experimental evaluation of course of this study. Because of either questionable benefits the fuel consumption and emissions reduction potential of low viscosity lubricants. SAE Paper 2009-01-1803. SAE International, Warrendale, or major implementation issues, it is highly uncertain Pa. whether any of these technologies will have any significant market penetration even in the 10- to 15-year time horizon.

OCR for page 38
57 SPARK-IGNITION GASOLINE ENGINES Hitomi, M., J. Sasaki, K. Hatamura, and Y. Yano. 1995. Mechanism of Schwarz, C., E. Schünemann, B. Durst, J. Fischer, and A. Witt. 2006. improving fuel efficiency by Miller-cycle and its future prospect. SAE Potentials of the spray-guided BMW DI combustion system. SAE Paper Paper 950974. SAE International, Warrendale, Pa. 2006-01-1265. SAE International, Warrendale, Pa. Hofmann, R., J. Liebl, M. Kluting, and R. Flierl. 2000. The new BMW 4-cyl- Sellnau, M., T. Kunz, J. Sinnamon, and J. Burkhard. 2006. 2-step variable inder gasoline engine—Reduction in fuel consumption without com- valve actuation: System optimization and integration on an SI engine. promise. SAE Paper 2001-08-0043. SAE International, Warrendale, Pa. SAE Paper 2006-01-0040. SAE International, Warrendale, Pa. Iwamoto, Y., K. Noma, O. Nakayama, and T. Yamauchi. 1997. Development Shahed, S.M., and Karl-Heinz Bauer. 2009. Parametric studies of the impact of gasoline direct injection engine. JSAE Technical Paper No. 9732009. of turbocharging on gasoline engine downsizing. SAE Paper 2009-01- Kojima, S. 2008. Development of high-performance and low-emission gaso- 1472. SAE International, Warrendale, Pa. line engine. SAE Paper 2008-01-0608. SAE International, Warrendale, Shin, S., A. Cusenza, and F. Shi. 2004. Offset crankshaft effects on SI engine Pa. combustion and friction performance. SAE Paper 2004-01-0606. SAE Kolwich, G. 2009. Light-Duty Technology Cost Analysis Pilot Study. Report International, Warrendale, Pa. FEV 07-069-103F. Prepared for U.S. Environmental Protection Agency, Sierra Research. 2008. Basic Analysis of the Cost and Long-Term Impact Ann Arbor, Mich. September 3. FEV, Inc., Auburn Hills, Mich. of the Energy Independence and Security Act Fuel Economy Standards. Kolwich, G. 2010. Light-Duty Technology Cost Analysis—Report on Sacramento, Calif., April 24. Additional Case Studies. Prepared for U.S. Environmental Protection Simko, A., M. Choma, and L. Repko. 1972. Exhaust emission control by the Agency. FEV, Inc., Auburn Hills, Mich. Ford-programmed combustion process—P ROCO. SAE Paper 720052. Kuwahara, K., T. Watanabe, H. Tanda, K. Ueda, and H. Ando. 2000. Intake- SAE International, Warrendale, Pa. port design for Mitsubishi GDI-Engine to realize distinctive in-cylinder Sorab, J., S. Korcek, C. Brower, and W. Hammer. 1996. Friction reducing flow and high charge coefficient. SAE Paper 2000-01-2801. SAE Inter- potential of low-viscosity-engine oils in bearings. SAE Paper 962033. national, Warrendale, Pa. SAE International, Warrendale, Pa. Leone, T.G., and M. Pozar. 2001. Fuel economy benefit of cylinder deacti- Stein, R.A., C.J. House, and T.G. Leone. 2009. Optimal use of E85 in a vation sensitivity to vehicle application and operating constraints. SAE turbocharged direct-injection engine. SAE Paper 2009-01-1490. SAE Paper 2001-01-3591. SAE International, Warrendale, Pa. International, Warrendale, Pa. Lindgren, R., M. Skogsberg, H. Sandquist, and I. Denbratt. 2003. The in- in- Sycomoreen. B MW Valvetronic. See http://sycomoreen.free.fr/imgs/ fluence of injector deposits on mixture formation in a DISC SI engine. IMAGES/autoconcept/ BMW_valvetronic.jpg. JSAE Technical Paper No. 20030110. Takemura, S., S. Aoyama, T. Sugiyama, T. Nohara, K. Moteki, M. Martec Group, Inc. 2008. Variable Costs of Fuel Economy Technologies. Nakamura, and S. Hara. 2001. A study of a continuous variable valve Prepared for Alliance of Automobile Manufacturers. June 1; amended event and lift (VEL) system. SAE Paper 2001-01-0243. SAE Interna- September 26 and December 10. tional, Warrendale, Pa. Misovec, K., B. Johnson, G. Mansouri, O. Sturman, and S. Massey. 1999. Tanaka, H., T. Nagashima, T. Sato, and S. Kawauchi. 1999. The effect of Digital valve technology applied to the control of an hydraulic valve 0W-20 low viscosity engine oil on fuel economy. SAE Paper 1999-01- actuator. SAE Paper 1999-01-0825. SAE International, Warrendale, Pa. 3468. SAE International, Warrendale, Pa. Nakada, M. 1994. Trends in engine technology and tribology. Triblogy Tanaka, Y., R. Hiyoshi, S. Takemura,Y. Ikeda, and M. Sugawara. 2007. A International 27(1), February. study of a compression-ratio-control mechanism for a multiple-link NESCCAF (Northeast States Center for a Clean Air Future). 2004. Reducing variable-compression-ratio engine. SAE Paper 2007-01-3547. SAE Greenhouse Gas Emissions from Light-Duty Motor Vehicles. March. International, Warrendale, Pa. NRC (National Research Council). 2002. Effectiveness and Impact of Cor- Taylor, C.M. 1998. Automobile engine tribology—Design considerations porate Average Fuel Economy (CAFE) Standards. National Academy for efficiency and durability. Wear 221:1-8. Press, Washington, D.C. Taylor, R.I., and C.R. Coy. 1999. Improved fuel efficiency by lubricant Petitjean, D., L. Bernardini, C. Middlemass, and S.M. Shahad. 2004. design: A review. Proceedings Institution of Mechanical Engineers Advanced gasoline engine turbocharging technology for fuel econ- 214, Part J. omy improvements. SAE Paper 2004-01-0988. SAE International, Toyota Motor Company. 2007. Toyota develops next generation engine Warrendale, Pa. valve mechanism. June 12, available at http://www2.toyota.co.jp/en/ Pischinger, S., K. Yapici, M. Schwanderlapp, and K. Habermann. 2001. news/07/0612.html. Variable compression in SI engines. SAE Paper 2001-24-0050. SAE Tuttle, J.H. 1980. Controlling engine load by means of late intake-valve International, Warrendale, Pa. closing. SAE Paper 800794. SAE International, Warrendale, Pa. Priest, M., and C.M. Taylor. 2000. Automobile engine tribology— Verhelst, S., J. De Landtsheere, F. De Smet, C. Billiouw, A. Trenson, and R. A pproaching the surface. Wear 241:193-203. SAE International, Sierens. 2008. Effects of supercharging, EGR and variable valve timing Warrendale, Pa. on power and emissions of hydrogen internal combustion engines. SAE Rebbert, M., G. Kreusen, and S. Lauer. 2008. A new cylinder deactivation Paper 2008-01-1033. SAE International, Warrendale, Pa. by FEV and Mahle. SAE Paper 2008-01-1354. SAE International, Wirbeleit, F., K. Binder, and D. Gwinner. 1990. Development of pistons Warrendale, Pa. with variable-compression-height for increasing efficiency and specific Ricardo, Inc. 2008. A Study of Potential Effectiveness of Carbon Dioxide power output of combustion engines. SAE Paper 900229. SAE Interna- Reducing Vehicle Technologies. Prepared for U.S. Environmental Pro- tional, Warrendale, Pa. tection Agency. EPA420-R-08-004. Contract No. EP-C-06-003. Work Wurms, R., M. Grigo, and W. Hatz. 2002. Audi FSI technology—Improved Assignment No. 1-14. Ann Arbor, Mich. performance and reduced fuel consumption. SAE Paper 2002-33-0002. Rosenberg, R.C. 1982. General friction considerations for engine design. SAE International, Warrendale, Pa. SAE Paper 821576. SAE International, Warrendale, Pa.

OCR for page 38
ANNEX 58 TABLE 4.A.1 Summary Table for Fuel Consumption Reduction Techniques for SI Engines: Incremental Percentage Reduction of Fuel Consumption with Associated Incremental Total Cost (with 1.5 RPE). See Figures 9.1 through 9.5 in Chapter 9 to understand the intended order for the incremental values. Consumption Benefit Incremental Cost $ Technologies I4 V6 V8 I4 V6 V8 Comments SI Techniques (%) Range (%) Range (%) Range Low High Low High Low High Low-viscosity • Small consumption benefit LUB 0.5 0.5 0.5 4.5 7.5 4.5 7.5 4.5 7.5 lubricants • Dependent on drive cycle • Roller follower valve trains and piston kit Engine friction EFR 0.5-2.0 0.5-2.0 1.0-2.0 48 78 72 117 96 156 friction reduction measures were nearly reduction universally implemented in the mid-1980s VVT—coupled • On SOHC setup cam phaser adjusts both cam phasing (CCP), CCP 1.5-3.0 1.5-3.5 2.0-4.0 52.5 105 105 exhaust and intake valve timing events SOHC • Manufacturer cost estimate of $35/phaser • Short durations may reduce pumping loss, and the reduced lift is a consequence of this • As intake manifold vacuum vanishes, alternate Discrete variable means must be found to implement power valve lift (DVVL), DVVL 1.5-3.0 1.5-3.0 2.0-3.0 195 240 270 315 420 480 brakes and PCV SOHC/DOHC • DVVL features two to three separate fixed profiles • Manufacturer cost estimate of $40/cylinder + $35/phaser ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES

OCR for page 38
Consumption Benefit Incremental Cost Technologies I4 V6 V8 I4 V6 V8 Comments SI Techniques (%) Range (%) Range (%) Range Low High Low High Low High • Effectiveness depends on power to weight ratio, previously added technologies, NVH, and drivability issues • Reduction in pumping losses from higher Cylinder cylinder loading DEAC NA 4.0-6.0 5.0-10.0 NA 510 600 536 630 deactivation, SOHC • Higher cost when applied to OHC engines • Manufacturer cost estimate for OHC engines of $340 to $400 • Additional manufacturer cost of $140 for NVH SPARK-IGNITION GASOLINE ENGINES issues • Effectiveness depends on power to weight ratio, previously added technologies, NVH, and drivability issues Cylinder DEAC NA 4.0-6.0 5.0-10.0 NA 330 375 383 • Reduction in pumping losses from higher deactivation, OHV cylinder loading • OHV has a lower cost when compared to OHC setups • Implementations include intake cam phaser (ICP) • Timing is important, and lift is merely a VVT—intake cam ICP 1.0-2.0 1.0-2.0 1.5-2.0 52.5 105 105 consequence of duration change phasing (ICP) • Some of this can be achieved with variable geometry intake manifolds • Manufacturer cost estimate of $35/phaser continued 59

OCR for page 38
TABLE 4.A.1 Continued 60 Consumption Benefit Incremental Cost Technologies I4 V6 V8 I4 V6 V8 Comments SI Techniques (%) Range (%) Range (%) Range Low High Low High Low High • Implementations include exhaust only and VVT—dual cam DCP 1.5-2.5 1.5-3.0 1.5-3.0 52.5 105 105 dual-cam phaser (DCP) phasing (DCP) • Manufacturer cost estimate of $35/phaser • Short durations may reduce pumping loss, and the reduced lift is a consequence of this • As intake manifold vacuum vanishes, Continuously alternate means must be found to implement variable valve lift CVVL 3.5-6.0 3.5-6.5 4.0-6.5 239 308 435 465 525 585 power brakes and PCV (CVVL) • CVVL features wide range of cam profiles • Manufacturer cost estimate of $300 for an I4, and $600 for a V-8 VVT—coupled cam • Requires in block cam phaser CCP 1.5-3.0 1.5-3.5 2.0-4.0 52.5 52.5 52.5 phasing (CCP), OHV • Manufacturer cost estimate of $35/phaser • Enables about +1.0 knock limited compression ratio • High pressure fuel pump increases parasitic loss Stoichiometric • Increased volumetric efficiency increases gasoline direct SGDI 1.5-3.0 1.5-3.0 1.5-3.0 176 293 254 384 443 527 pumping loss injection (GDI) • Injector deposits formed upon hot shut down has been a traditional concern • Manufacturer cost estimates $80/cylinder and $136 for injector noise abatement items Consumption Benefit Incremental Cost Technologies Comments I4 V6 V8 I4 V6 V8 SI Techniques (%) Range (%) Range (%) Range Low High Low High Low High • Vehicle launch performance will likely be compromised • Piston underside oil squirters, an oil cooler, and an intercooler may contribute to system Turbocharging and TRBDS 2.0-5.0 4.0-6.0 4.0-6.0 555 735 -50 308 788 1185 merits downsizing • Dual scroll and VNT units will improve vehicle launch performance • Manufacturer estimates $550-$920 for a fixed geometry system ASSESSMENT OF FUEL ECONOMY TECHNOLOGIES FOR LIGHT-DUTY VEHICLES