|
Items in cart [0] |
Questions? Call 888-624-8373 |
| [ Top of Page ] [ Home ] [ Contact Us ] [ Help ] [ The National Academies Home ] | ||
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 31
Technologies for Improving the Fuel Economy of
Passenger Cars and LighI-Duty Trucks
This chapter examines a variety of technologies that could
be applied to improve the fuel economy of future passenger
vehicles. It assesses their fuel economy potential, recogniz-
ing the constraints imposed by vehicle performance, func-
tionality, safety, cost, and exhaust emissions regulations.
The committee reviewed many sources of information
related to fuel economy-improving technologies and their
associated costs, including presentations at public meet-
ings and available studies and reports. It also met with au-
tomotive manufacturers and suppliers and used consultants
to provide additional technical and cost information (EEA,
2001; Sierra Research, 2001~. Within the time constraints
of this study, the committee used its expertise and engi-
neering judgment, supplemented by the sources of infor-
mation identified above, to derive its own estimates of the
potential for fuel economy improvement and the associ-
ated range of costs.
In addition, after the prepublication copy of the report
was released in July 2001, the committee reexamined its
technical analysis. Representatives of industry and other
groups involved in fuel efficiency analysis were invited to
critique the committee's methodology and results. Several
minor errors discovered during this reexamination have been
corrected in this chapter, and the discussion of the methodol-
ogy and results has been clarified. The reexamination is pre-
sented in Appendix F.
FUEL ECONOMY OVERVIEW
To understand how the fuel economy of passenger ve-
hicles can be increased, one must consider the vehicle as a
system. High fuel economy is only one of many vehicle
attributes that may be desirable to consumers. Vehicle per-
formance, handling, safety, comfort, reliability, passenger-
and load-carrying capacity, size, styling, quietness, and
costs are also important features. Governmental regulations
require vehicles to meet increasingly stringent require-
3
7
meets, such as reduced exhaust emissions and enhanced
safety features. Ultimately these requirements influence
final vehicle design, technology content, and the subject
of this report fuel economy. Manufacturers must assess
trade-offs among these sometimes-conflicting characteris-
tics to produce vehicles that consumers find appealing and
affordable.
Engines that burn gasoline or diesel fuel propel almost all
passenger cars and light-duty trucks. About two-thirds of the
available energy in the fuel is rejected as heat in the exhaust
and coolant or frictional losses. ~ The remainder is transformed
into mechanical energy, or work. Some of the work is used to
overcome frictional losses in the transmission and other parts
of the drive train and to operate the vehicle accessories (air
conditioning, alternator/generator, and so on). In addition,
standby losses occur to overcome engine friction and cooling
when the engine is idling or the vehicle is decelerating.
As a result, only about 12 to 20 percent of the original
energy contained in the fuel is actually used to propel the
vehicle. This propulsion energy overcomes (1) inertia
(weight) when accelerating or climbing hills, (2) the resis-
tance of the air to the vehicle motion (aerodynamic drag),
and (3) the rolling resistance of the tires on the road. Con-
sequently, there are two general ways to reduce vehicle
fuel consumption: (1) increase the overall efficiency of the
powertrain (engine, transmission, final drive) in order to
deliver more work from the fuel consumed or (2) reduce
the required work (weight, aerodynamics, rolling resis-
iTheoretically gasoline or diesel engines (and fuel cells) can convert all
of the fuel energy into useful work. In practice, because of heat transfer,
friction, type of load control, accessories required for engine operation,
passenger comfort, etc., the fraction used to propel the vehicle varies from
as low as zero (at idling) to as high as 40 to 50 percent for an efficient diesel
engine (gasoline engines are less efficient). Further losses occur in the drive
train. As a result, the average fraction of the fuel converted to work to
propel the vehicle over typical varying-load operation is about 20 percent of
the fuel energy.
OCR for page 32
32
ENGINE
ACCESSORIES
TRANSMISSION
FINAL DRIVE
SYSTEM EFF ~ FM— ~
FUEL CONSUMPTION
FIGURE 3-1 Energy use in vehicles. SOURCE: Adapted from Riley (1994~.
lance, and accessory load) to propel the vehicle. These con-
cepts are illustrated in Figures 3-1 and 3-2. Regenerative
braking and shutting the engine off during idling also save
energy, as discussed in the section on hybrid electric ve-
hicles, below.
Vehicle fuel economy currently is determined according
to procedures established by the Environmental Protection
Agency (EPA). Vehicles are driven on a dynamometer in a
controlled laboratory (in order to eliminate weather and road
variables. Both city and highway driving are simulated.
The city test is a 7.5-mile trip lasting 23 minutes with 18
stops, at an average speed of about 20 miles per hour (mph).
About 4 minutes are spent idling (as at a traffic light), and a
short freeway segment is included. The vehicle begins the
test after being parked overnight at about 72°F (22°C) (cold
soak). The highway test is a 10-mile trip with an average
speed of about 48 mph. The test is initiated with a warmed-
up vehicle (following the city test) and is conducted with no
stops and very little idling. The basis for compliance with
CAFE (and comparison of the technologies below) is the
current EPA Federal Test Procedure (FTP-75) with city,
Aerodynamic drag is accounted for in the results by incorporating coast-
down data from other tests. Nevertheless, there are significant differences
between the mileage tests and real-life driving. For example, the dynamom-
eter is connected to only one pair of tires, but on the road, all tires are
rolling. Most drivers experience lower fuel economy than suggested by
EPA's results. It should be noted that the test driving cycles were derived
from traffic pattern observations made many years ago, which may not be
representative now. A review of the validity of the test cycles for today's
traffic patterns would seem appropriate.
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
INERTIA (WEIGHT)
ROLLING RESISTANCE |
AERODYNAMIC DRAG
ROAD LOAD
highway, and combined (55 percent city/45 percent high-
way) ratings in miles per gallon (mpg) (CFR, 2000).
During city driving, conditions such as acceleration, en-
gine loading, and time spent braking or at idle are continu-
ally changing across a wide range of conditions. These varia-
tions result in wide swings in fuel consumption. Inertial loads
and rolling resistance (both directly related to weight) com-
bined account for over 80 percent of the work required to
move the vehicle over the city cycle, but less for the high-
way cycle. A reduction in vehicle weight (mass) therefore
has a very significant effect on fuel consumption in city driv-
ing. This strong dependence on total vehicle weight explains
why fuel consumption for the new vehicle fleet correlates
linearly with vehicle curb weight, as shown in Attachment
3A.
Weight reduction provides an effective method to reduce
fuel consumption of cars and trucks and is an important goal
for the government-industry program Partnership for a New
Generation of Vehicles (PNGV). Reducing the required pro-
pulsion work reduces the load required from the engine, al-
lowing the use of a smaller engine for the same performance.
In the search for lightweight materials, PNGV has focused
on materials substantially lighter than the steel used in most
current vehicles. Components and body structure fabricated
from aluminum, glass-fiber-reinforced polymer composites
(GFRP), and carbon-fiber-reinforced polymer composites
(CFRP), including hybrid structures, are being investigated
(NRC, 2000a).
Reducing vehicle weight without reducing practical space
for passengers and cargo involves three strategies: (1) sub-
stitution of lighter-weight materials without compromising
OCR for page 33
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
' Exhaust heat
Fuel
Energy
l
l
Indicated
Work
l
>
Cooling system
~ _
', Engine friction At>
, Pumping losses>
Engine
Output
(Brake work)
1 1
1 1
1 1
'1 1
~ Accessories
. _
1
, Transmission ~
i Inertia
Vehicle L ~
~ _
, Rolling resistance
33
FIGURE 3-2 Where the energy in the fuel goes (proportions vary with vehicle design, type of engine, and operating conditions). SOURCE:
NRC (1992~.
structural strength (e.g., aluminum or plastic for steel);
(2) improvement of packaging efficiency, that is, redesign of
the drive train or interior space to eliminate wasted space;
and (3) technological change that eliminates equipment or
reduces its size. Design efficiency and effectiveness can also
result in lighter vehicles using the same materials and the
same space for passengers and cargo.
Automotive manufacturers must optimize the vehicle and
its powertrain to meet the sometimes-conflicting demands
of customer-desired performance, fuel economy goals, emis-
sions standards, safety requirements, and vehicle cost within
the broad range of operating conditions under which the ve-
hicle will be used. This necessitates a vehicle systems analy-
sis. Vehicle designs trade off styling features, passenger
value, trunk space (or exterior cargo space for pickups), and
utility. These trade-offs will likewise influence vehicle
weight, frontal area, drag coefficients, and power train pack-
aging, for example. These features, together with engine per-
formance, torque curve, transmission characteristics, control
system calibration, noise control measures, suspension char-
acteristics, and many other factors, will define the drivability,
customer acceptance, and marketability of the vehicle.
Technology changes modify the system and hence have
complex effects that are difficult to capture and analyze. It is
usually possible, however, to estimate the impacts of spe-
cific technologies in terms of a percentage savings in fuel
consumption for a typical vehicle without a full examination
of all the system-level effects. Such a comparative approach
is used in this chapter.3
Although CAFE standards and EPA fuel economy rat-
ings are defined in the now-familiar term miles per gallon
(mpg), additional assessment parameters have been identi-
3Further explanation of the methodology is provided in Appendix F.
fled to assist in the evaluation process, including fuel con-
sumption in gallons per 100 miles; load-specific fuel
consumption (LSFC) in gallons per ton (of cargo plus pas-
sengers) per 100 miles; and weight-specific fuel consump-
tion (WSFC) in gallons per vehicle weight per 100 miles.
Attachment 3A further explains why these parameters are
meaningful engineering relationships by which to judge fuel
economy and the efficiency of moving the vehicle and its
intended payload over the EPA cycle.
Figure 3-3 shows the actual energy efficiency of vehicles
of different weights. For both city and highway cycles, the
fuel consumed per ton of weight and per 100 miles is plotted
against the weight of the vehicle. Normalizing the fuel con-
sumption (dividing by weight) yields an efficiency factor (in
an engineering sense), which is particularly useful in com-
paring fuel savings opportunities. It is also useful that the
weight-adjusted or normalized values can be reasonably ap-
proximated by a horizontal line. Points above the line repre-
sent vehicles with lower-than-average efficiency, which re-
quire more than the average amount of fuel to move a vehicle
of a given weight over the EPA certification cycle. In prin-
ciple, given sufficient lead time and business incentives (eco-
nomic or regulatory), the vehicles above the line could be
improved to the level of those below the line, within the
limits of customer-desired performance and vehicle utility
features. As an example, a 4,000-lb vehicle, in principle,
could drop from 2 gallons per ton- 100 miles (25 mpg) on the
highway cycle to 1.4 (35.7 mpg) using technologies dis-
cussed later in this chapter. However, although larger,
heavier vehicles have greater fuel consumption than smaller,
lighter vehicles, their energy efficiencies in moving the ve-
hicle mass (weight) are very similar. These data also suggest
that the potential exists to improve fuel consumption in fu-
ture vehicles. However, changing conditions such as safety
OCR for page 34
OCR for page 35
OCR for page 36
OCR for page 37
OCR for page 38
OCR for page 39
OCR for page 40
OCR for page 41
OCR for page 52
OCR for page 53
OCR for page 54
OCR for page 55
OCR for page 56
OCR for page 57
OCR for page 58
OCR for page 59
OCR for page 60
OCR for page 61
OCR for page 62
Representative terms from entire chapter:
fuel economy
34
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
A.=
2.8
In
a)
° 2.4
l
en
. _
~ 2.0
a)
.O
> 1.6
o
~ 1.2-
o
ct
0.8
0.4 -
8
1600 2000 2400 2800 3200
3600 4000 4400
Weight, lb
4800 5200 5600 6000 6400 6800
FIGURE 3-3 EPA data for fuel economy for MY 2000 and 2001 cars and light trucks. SOURCE: EPA, available online at
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
that will be available in 2006 (EPA, 2000b), has delayed
production decisions. In general, the committee believes that
the Tier 2 NOX and PM standards will inhibit, or possibly
preclude, the introduction of diesels into vehicles under
8,500 lb unless cost-effective, reliable, and regulatory-com-
pliant exhaust gas aftertreatment technology develops rap-
idly. A key challenge is the development of emission control
systems that can be certified for a 120,000-mile lifetime.
In theory, the bin system will allow diesels to penetrate
the light-duty vehicle market, but manufacturers must still
meet the stringent fleet average standard. For example, for
every vehicle in bin 8 (0.2 g/mile NOX), approximately seven
vehicles in bin 3 (0.03 g/mile) would have to be sold in order
to meet the 0.07 g/mile fleet-average NOX standard.
These same factors have caused the committee to con-
clude that major market penetration of gasoline direct-injec-
tion engines that operate under lean-burn combustion, which
is another emerging technology for improving fuel economy,
is unlikely without major emissions-control advancements.
California's exhaust emission requirements super
ultralow emission vehicle (SULEV) and partial zero emis-
sion vehicle (PZEV) are also extremely challenging for the
introduction of diesel engines. In particular, the California
Air Resources Board (CARB) has classified PM emissions
from diesel-fueled engines as a toxic air contaminant
(CARE, 1998~. (Substances classified as toxic are required
to be controlled.)
TECHNOLOGIES FOR BETTER FUEL ECONOMY
The 1992 NRC report outlined various automotive tech-
nologies that were either entering production at the time or
were considered as emerging, based on their potential and
production intent (NRC, 1992~. Since then, many regulatory
and economic conditions have changed. In addition, auto-
motive technology has continued to advance, especially in
microelectronics, mechatronics, sensors, control systems,
and manufacturing processes. Many of the technologies
identified in the 1992 report as proven or emerging have
already entered production.
The committee conducted an updated assessment of
various technologies that have potential for improving fuel
economy in light-duty vehicles. This assessment takes
into account not only the benefits and costs of applying the
technologies, but also changes in the economic and regula-
tory conditions, anticipated exhaust emission regulations,
predicted trends in fuel prices, and reported customer
preferences.
The technologies reviewed here are already in use in some
vehicles or are likely to be introduced in European and
Japanese vehicles within 15 years. They are discussed belong
under three general headings: engine technologies, transmis-
sion technologies, and vehicle technologies. They are listed
in general order of ease of implementation or maturity of the
technology (characterized as "production intent" or "emerg-
35
ing"~. The committee concludes its assessment of potential
technologies with some detailed discussion of the current
and future generations of hybrid vehicles and fuel- cell power
sources.
For each technology assessed, the committee estimated
not only the incremental percentage improvement in fuel
consumption (which can be converted to fuel economy in
miles per gallon [mpg] to allow comparison with current
EPA mileage ratings) but also the incremental cost that ap-
plying the technology would add to the retail price of a ve-
hicle. The next subsection of this chapter, "Technologies
Assessed," reviews the technologies and their general ben-
efits and challenges.
After that, the section "Estimating Potential Fuel Econ-
omy Gains and Costs" presents estimates of the fuel con-
sumption benefits and associated retail costs of applying
combinations of these technologies in 10 classes of produc-
tion vehicles. For each class of vehicle, the committee
hypothesizes three exemplary technology paths (technology
scenarios leading to successively greater improvements in
fuel consumption and greater cost).
Technologies Assessecl
The engine, transmission, and vehicle technologies dis-
cussed in this section are all considered likely to be available
within the next 15 years. Some (called "production intent" in
this discussion) are already available, are well known to
manufacturers and their suppliers, and could be incorporated
in vehicles once a decision is reached to use them. Others
(called "emerging" in this discussion) are generally beyond
the research phase and are under development. They are suf-
ficiently well understood that they should be available within
10 to 15 years.
Engine Technologies
The engine technologies discussed here improve the
energy efficiency of engines by reducing friction and other
mechanical losses or by improving the processing and com-
bustion of fuel and air.
Production-Intent Engine Technologies The engine tech-
nologies discussed here could be readily applied to produc-
tion vehicles once a decision is made to proceed, although
various constraints may limit the rate at which they penetrate
the new vehicle fleet:
Engine friction and other mechanical/hydrodynamic
loss reduction. Continued improvement in engine
component and system design, development, and com-
puter-aided engineering (CAB) tools offers the poten-
tial for continued reductions of component weight and
thermal management and hydrodynamic systems that
improve overall brake-specific efficiency. An im-
36
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
.
.
.
.
.
provement in fuel consumption of 1 to 5 percent is
considered possible, depending on the state of the
baseline engine.
Application of advanced, low-friction lubricants. The
use of low-friction, multiviscosity engine oils and
transmission fluids has demonstrated the potential to
reduce fuel consumption by about 1 percent, compared
with conventional lubricants.
Multivalve, overhead camshaft valve trains. The ap-
plication of single and double overhead cam designs,
with two, three, or four valves per cylinder, offers the
potential for reduced frictional losses (reduced mass
and roller followers), higher specific power (tap/liter),
engine downsizing, somewhat increased compression
ratios, and reduced pumping losses. Depending on the
particular application and the trade-offs between valve
number, cost, and cam configuration (single overhead
cam [SOHC] or double overhead cam [DOHC]), im-
provements in fuel consumption of 2 to 5 percent are
possible, at constant performance, including engine
downsizing (Chon and Heywood, 2000~. However,
market trends have many times shown the use of these
concepts to gain performance at constant displace-
ment, so that overall improvements in fuel consump-
tion may be less.
Variable valve timing (AFT). Variation in the cam
phasing of intake valves has gained increasing market
penetration, with an associated reduction in produc-
tion cost. Earlier opening under low-load conditions
reduces pumping work. Under high-load, high-speed
conditions, variations in cam phasing can improve
volumetric efficiency (breathing) and help control re-
sidual gases, for improved power. Improvements in
fuel consumption of 2 to 3 percent are possible through
this technology (Chon and Heywood, 2000; Leone et
al., 1996~.
Variable valve lift and timing (VOLT). Additional ben-
efits in air/fuel mixing, reduction in pumping losses,
and further increases in volumetric efficiency can be
gained through varying timing and valve lift (staged or
continuous). Depending upon the type of timing and
lift control, additional reductions in fuel consumption
of 1 to 2 percent, above cam phasing only, are possible
(Pierik and Burkhanrd, 2000), or about 5 to 10 percent
compared to two-valve engines (including downsizing
with constant performance).
Cylinder deactivation. An additional feature that can
be added to variable valve lift mechanisms is to allow
the valves of selected cylinders to remain closed, with
the port fuel injection interrupted. Currently, this tech-
nology is applied to rather large engines (>4.0 liter) in
V8 and V12 configurations. This approach, which is
sometimes referred to as a variable displacement en-
gine, creates an "air spring" within the cylinder. A1-
though both frictional and thermodynamic losses oc-
.
cur, they are more than offset by the increased load
and reduced specific fuel consumption of the remain-
ing cylinders. However, engine transient performance,
idle quality, noise, and vibration can limit efficiency
gains and must be addressed. Improvements in fuel
consumption in the range of 3 to 6 percent are pos-
sible, even given that reductions in throttling losses
associated with higher load factors over the operating
cycle cannot be double counted.
Engine accessory improvement. As engine load and
speed ranges continue to advance, many engine acces-
sories such as lubrication and cooling systems and
power steering pumps are being optimized for reduc-
tions in energy consumption and improved matching
of functionality over the operating range. The evolu-
tion of higher-voltage (i.e., 42 V) powertrain and ve-
hicle electrical systems will facilitate the cost-efficient
applications of such components and systems. Im-
provements in fuel consumption of about 1 to 2 per-
cent are possible with such technologies.
Engine downsizing and supercharging. Additional im-
provements in fuel consumption can be gained by re-
ducing engine displacement and increasing specific
power (while maintaining equal performance) by
boosting the engine (turbocharger or mechanical su-
percharger). Degraded transient performance (turbo-
lag) typically associated with turbochargers can be sig-
nificantly offset by incorporating variable geometry
turbines or mechanical (positive displacement) super-
chargers. Additional modifications for transmission
matching, aftertreatment system warm-up, and other
factors that can degrade exhaust emissions control
must also be considered. Improvements in fuel con-
sumption of 5 to 7 percent are considered possible with
this approach, at equivalent vehicle performance
(Ecker, 2000~. However, when this concept is com-
bined with multivalve technology, total improvements
of about 10 percent are possible compared with a two-
valve engine baseline.
Emerging Engine Technologies The following engine tech-
nologies are considered emerging for passenger car and light-
duty truck applications. Significant market penetration in the
United States is likely to take 5 to 10 years. Some of them
are already in production elsewhere (in Japan or Europe),
where they may benefit from high fuel taxes, government
incentives for particular engine types or displacements, and
more lenient exhaust emission or vehicle safety standards.
The discussion that follows outlines not only the benefits but
also the technical challenges or economic hurdles for each
technology.
Intake valve throttling (IVT). Advances in micropro-
cessor technology, feedback control, electromechani-
cal actuation, sensor technology, and materials con-
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
.
.
tinue to accelerate. As a result, electromechanical
IVT is advancing to the point where BMW has an-
nounced the introduction of its so-called Valvetronic
concept. When multipoint fuel injection is used, both
the lift and timing of the intake valves can be con-
trolled to maintain the correct air/fuel ratio without a
throttle plate. This has the potential to essentially
eliminate the pumping losses across the normal but-
terfly throttle valve. Also important is the potential
to use conventional three-way-catalyst (TWC) after-
treatment technology and incorporate cylinder deac-
tivation. However, significant cost and complexity in
actuation, electronic control, and system calibration
are to be expected. Improvements in fuel consump-
tion of an additional 3 to 6 percent above VVLT are
possible with this technology. Compared with two-
valve engines, total system improvements may ap-
proach 6 to 12 percent.
Camless valve actuation (CVAJ. A further evolution
of fast-acting, completely variable valve timing (not
limited by the lift curve of a camshaft) is represented
by electromechanical solenoid-controlled, spring-
mass valve (EMV) systems (Siemens, BMW, FEV)
and high-pressure hydraulic-actuated valves with
high-speed, digital control valve technology (Ford
Navistar). In addition to reducing pumping losses,
this technology facilitates intake port and cylinder
deactivation and allows the use of conventional
TWC aftertreatment. Technical challenges in the
past for EMV have been to minimize energy con-
sumption and achieve a soft landing of the valve
against the seat during idle and low-speed, low-load
operation, for acceptable noise levels. These issues
appear to be solved through advances in sensor and
electromagnetic technologies. EMV systems are ex-
pected to see limited production within 5 to 7 years.
Improvements in fuel consumption of 5 to 10 per-
cent relative to VVTL are possible with this tech-
nology. Compared with fixed-timing, four-valve
engines, total system improvements of 15 percent
or more have been demonstrated (Pischinger et al.,
2000~.
Variable compression ratio (VCRJ engines. Current
production engines are typically limited in compres-
sion ratio (CR) to about 10:1 to 10.5:1 with the use of
high-octane fuel, owing to knocking under high load.
However, significant improvements in fuel consump-
tion could be gained with higher CR under normal
driving cycles. Many different VCR approaches that
allow improved efficiency under low load with high
CR (13-14:1) and sufficient knock tolerance under full
load with lower CR (~8:1) are under development.
Saab appears to have the most advanced VCR proto-
types. Automakers, suppliers, and R&D organizations
are currently exploring many other approaches that are
37
applicable to both inline and Vee engine configura-
tions. Several of these are expected to enter production
within 10 years. Compared with a conventional four-
valve VVT engine, improvements in fuel consumption
of 2 to 6 percent are possible (Wirbeleit et al., 1990~.
The combination of VCR with a supercharged,
downsized engine is likely to be effective, giving the
maximum advantage of both systems and reducing
total fuel consumption, at constant performance, by
10 to 15 percent. However, the potential complexity
of the hardware, system durability, control system
development, and cost must be traded off for produc-
tion applications.
Many additional engine technologies with good potential
for improved fuel consumption are the subject of R&D.
Others are currently offered in markets with higher fuel
prices (due to higher taxes) or exhaust emission standards
more lenient than the upcoming federal Tier 2 emission stan-
dards (or California's SULEV standards, set to begin in
model year 2004~. A brief summary of these technologies is
presented below, including reference to the areas of uncer-
tainty and the need for further development.
Direct-injection (DIJ gasoline engines. Stratified-
charge gasoline engines burning in a lean mode (when
more air is present than required to burn the fuel) offer
improved thermodynamic efficiency. However, the
technology faces potential problems in controlling par-
ticulate emissions and NOX. Trade-offs between the
maximum operating range under lean conditions ver-
sus stoichiometric operation (when the exact amount
of air needed to burn the fuel is present) with early
injection must be developed. Although lean-burn DI
engines of the type offered in Europe could improve
fuel consumption by more than 10 percent, NOx-con-
trol requirements that necessitate stoichiometric op-
eration and the use of TWCs limit the potential fuel
consumption improvement to between 4 and 6 percent
(Zhao and Lai, 1997~.
Direct-injection diesel engines. The application of
small (1.7- to 4.0-liter), high-speed (4,500-rpm), tur-
bocharged, direct-injection diesel engines has seen tre-
mendous expansion in passenger cars and light-duty
trucks in Europe. Increasing power densities (>70 hp/
liter), achieved through the application of advanced,
high-pressure, common-rail fuel injection systems;
variable geometry turbochargers; and advances in
noise, vibration, and harshness (NVH) control tech-
nologies, combined with high-efficiency, lean-burn
combustion systems and practically smokeless and
odorless emissions, have greatly improved customer
acceptance in Europe. The high low-speed torque and
relatively flat torque curve also offer significant
drivability improvements. Fuel consumption improve-
38
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
meets of 30 to 40 percent or more are possible com-
pared with conventional two-valve gasoline engines.
The challenges, which inhibit widespread introduction
in the United States, include meeting strict NOX and
particulate emission standards for Tier 2 and SULEV;
much higher engine and vehicle purchase price ($2,000
to $3,000) than conventional gasoline engines; and
uncertain U.S. customer acceptance. The creation of
NOX and particulate emissions is exacerbated by the
stratification present in the fuel/air mixture resulting
from in-cylinder injection. R&D activities continue on
emission control through advanced combustion and
fuel injection concepts, fuel composition (low sulfur),
aftertreatment technologies (selective catalytic reduc-
tion [SCR], NOX traps, particulate filters), and control
of noise and vibration. Although DI diesel engines are
offered in some trucks over 8,500 lb and are offered by
one manufacturer (VW) for passenger cars under cur-
rent Tier 1 emissions standards, wide use in sport util-
ity vehicles (SUVs) and light-duty trucks has not
occurred, and the ability of this technology to comply
with the upcoming Tier 2 and SULEV standards is
highly uncertain.
Transmission Technologies
The second group of technologies assessed by the com-
m~ttee involves improvements in the efficiency with which
power is transmitted from the engine to the driveshaft or
axle.
Production-intent Transmission Technologies Over the
past 10 years, transmission technologies have been evolving
toward increasing electronic control, adapting torque con-
verter lock-up, four- and five-speed automatics (from three-
and four-speed), and various versions of all-wheel drive
(AWD) or four-wheel drive (4WD) and traction control,
ranging from continuous, traction-controlled AWD to auto-
matic 2WD-4WD traction control in some SUVs.
.
.
approaches are also being pursued for future produc-
tion. Depending on the type of CVT and the power/
speed range of the engine, this technology can improve
fuel consumption by about 4 to 8 percent. However,
production cost, torque limitations, and customer ac-
ceptance of the system's operational characteristics
must be addressed.
Emerging Transmission Technologies Automotive manu-
facturers continue to seek ways to reduce the mechanical
(frictional and hydrodynamic) losses of transmissions and
improve their mating with engines. The various types of hy-
brid vehicles will also involve changes in conventional trans-
missions. These emerging technologies are likely to be avail-
able in the latter part of the current decade.
.
.
Five-speed automatic transmission. A five-speed au-
tomatic transmission permits the engine to operate in
its most efficient range more of the time than does a
four-speed transmission. A fuel consumption improve-
ment of 2 to 3 percent is possible, at constant vehicle
performance, relative to a four-speed automatic.
Continuously variable transmission (CVT). Several
versions of continuously variable transmissions are
offered in production in Europe and Japan and a few in
the United States (by Honda and Toyota). Historically,
these transmission types have used belts or chains of
some kind to vary speed ratios across two variable-
diameter pulleys. The major production units utilize
compression belts (VanDorne) or tension chains. Other
Automatic transmission with aggressive shift logic.
Shift schedules, logic, and control of torque transfer
can significantly affect perceived shift quality. Ad-
vanced work on methods to reduce losses associated
with torque converters or torque dropout is being pur-
sued. It is estimated that a 1 to 3 percent improvement
in fuel consumption can be obtained through such
measures. However, these will be highly affected by
customer perception in the United States and may re-
quire quite some time for significant acceptance.
Six-speed automatic transmission. Advanced six-
speed automatic transmissions can approach the per-
formance of CVT transmissions without limitations in
the ability to transmit torque. An additional improve-
ment of 1 to 2 percent in fuel consumption is possible,
compared to a five-speed automatic. Based on their
higher cost and control complexity, such transmissions
will probably see only limited introduction namely,
in high-end luxury or performance vehicles.
Automatic shift/manual transmission (ASM/AMT). In
the continuing quest to reduce mechanical losses,
manufacturers are developing new generations of au-
tomatic transmissions that eliminate the hydraulic
torque converter and its associated pump, replacing it
with electronically controlled clutch mechanisms. This
approach offers two basic possibilities: The torque
from different gear sets can be intermittently inter-
rupted (as in a conventional manual transmission)
through the use of a single electronically controlled
clutch; or the torque can be continuously controlled,
without dropout, through the use of two electronically
controlled clutches. Improvements in fuel consump-
tion of 3 to 5 percent over a conventional four-speed
automatic transmission with hydraulic torque con-
verter are possible. However, increased cost, control
system complexity, durability, and realizable fuel con-
sumption gain versus acceptable shift quality for U.S.
customers must be addressed.
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
· Advanced CVT. Continued advances in methods for
high-efficiency, high-torque transfer capability of
CVTs are being pursued. New versions of CVTs that
will soon enter production incorporate toroidal fric-
tion elements or cone-and-ring assemblies with vary-
ing diameters. However, these versions also have
trade-offs of torque capability vs. frictional losses.
These next-generation transmissions have the poten-
tial to improve fuel consumption by about O to 2 per-
cent (relative to current CVTs), with higher torque ca-
pabilities for broader market penetration. However,
production cost, system efficiency, and customer ac-
ceptance of the powertrain operational characteristics
must still be addressed.
Vehicle Technologies
By reducing drag, rolling resistance, and weight, the fuel
consumption of vehicles could, in principle, be cut rather
sharply in the relatively near term. Manufacturers, however,
would quickly run into serious trade-offs with performance,
carrying capacity, and safety. Also to be considered are novel
vehicle concepts such as hybrid electrics, powered by vari-
ous combinations of internal combustion engines and batter-
ies or fuel cells. The following discussion reviews both
production-intent and emerging vehicle technologies.
Production-Intent Vehicle Technologies The following
fuel consumption measures are deemed available in the near
term (almost immediately after a decision to use them is
made):
.
.
Aerodynamic drag reduction on vehicle designs. This
improvement can be very cost-effective if incorporated
during vehicle development or upgrades. However,
vehicle styling and crashworthiness have significant
influences on the ultimate levels that can be achieved.
For a 10 percent reduction in aerodynamic drag, an
improvement in fuel consumption of 1 to 2 percent can
be achieved. As drag coefficients proceed below about
0.30, however, the design flexibility becomes limited
and the relative cost of the vehicle can increase dra-
matically. Substituting video minicameras for side-
view mirrors (e.g., as is being done for the PNGV con-
cept vehicles) would be advantageous but would
necessitate a change in safety regulations (NRC,
2000a).
Rolling resistance. Continued advances in tire and
wheel technologies are directed toward reducing roll-
ing resistance without compromising handling, com-
fort, or braking. Improvements of about 1 to 1.5
percent are considered possible. The impacts on per-
formance, comfort, durability, and safety must be
evaluated, however.
39
Vehicle weight reduction. Reducing vehicle weight
while maintaining acceptable safety is a difficult bal-
ance to define. While most manufacturers believe that
some reduction in vehicle weight can be accom-
plished without a measurable influence on in-use
safety, debate continues on how much weight can be
reduced without compromising crush space, by using
lighter-weight materials and better, more crashwor-
thy designs.
Emerging Vehicle Technologies Several advanced vehicle
technologies are being considered for near-term production.
Interest in these technologies has been fostered by the PNGV
program. In addition, a wide variety of hybrid vehicle tech-
nologies are being explored for initial introduction within
the next 5 to 10 years. This section reviews vehicle technolo-
gies that have been identified by the industry for introduc-
tion within the next 10 years.
.
.
.
Forty-two volt electrical system. Most automotive
manufacturers are planning a transition to 42-V elec-
trical systems to support the continuing need for in-
creased electrical power requirements for next-genera-
tion passenger vehicles. Higher voltage will reduce
electrical losses and improve the efficiency of many
onboard electrically powered systems. It will also al-
low new technologies such as electric power steering,
which can be significantly more efficient than current
technology. Fuel consumption reductions associated
with the implementation and optimization of related
systems are expected to range from 1 to 2 percent.
Integrated starter/generator (ISG). Significant im-
provements in fuel consumption under real-world op-
erating conditions can be gained by turning the engine
off during idle, while operating the necessary acces-
sories electrically (air conditioning presents a major
challenge, however). ISG systems providing nearly in-
stantaneous engine restart are now planned for pro-
duction. Idle stop, under many conditions, is expected
to achieve a 4 to 7 percent reduction in fuel consump-
tion. Depending on the size and type of battery chosen,
it is also possible to recover electrical energy through
regenerative braking and subsequent launch assist us-
ing ISG technology. Doing so adds cost, complexity,
and weight but could improve fuel consumption by a
total of 5 to 10 percent.
Hybrid electric vehicles. Hybrid electric vehicles of
various types are in different stages: Some are starting
to be introduced, others are in advanced stages of de-
velopment, and still others are the focus of extensive
research by nearly all the large automotive manufac-
turers. They include so-called "mild hybrids" (with
regenerative braking, ISG, launch assist, and minimal
battery storage); "parallel hybrids" (with the engine
40
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
powering either or both a mechanical drive train and
an electric motor/generator serving as additional pro-
pulsion to recharge the battery); and "series hybrids"
(in which the engine does not drive the wheels but
always drives an electric motor/generator to propel the
vehicle, recharge the battery, or perform both func-
tions simultaneously).
The method and extent of hybridization depends on
the vehicle type, its anticipated use, accessory pack-
age, type of battery, and other considerations. The an-
ticipated improvements in fuel consumption can there-
fore vary, from about 15 percent for certain mild
hybrids to about 30 percent for parallel hybrids. In gen-
eral, series hybrids are not yet intended for even lim-
ited production, owing to the relatively poor perfor-
mance of electric power propulsion and the low
efficiencies of current battery systems compared with
mechanical drive systems. The varying complexity of
the different hybrid types is reflected in large varia-
tions in incremental cost. The cost premium of today's
limited-production mild hybrids is predicted to be
$3,000 to $5,000 when they reach production volumes
over 100,000 units per year. For fully parallel systems,
which operate for significant periods entirely on the
electrical drive, especially in city driving, the cost pre-
mium can escalate to $7,500 or more. In addition to
offering significant gains in fuel consumption, these
vehicles have the potential for beneficial impacts on
air quality. Further information on hybrids is provided
in a separate section below.
· Fuel-cell hybrid electric vehicles. The most advanced
emerging vehicle technology currently under research
and development substitutes an electrochemical fuel
cell for the internal combustion engine. In proton ex-
change membrane (PEM) fuel cells, hydrogen and
oxygen react to produce electricity and water. Since
gaseous hydrogen is difficult to store with reasonable
energy density, many manufacturers are pursuing the
decomposition of a liquid fuel (either methanol or
gasoline) as a source of hydrogen, depending on cor-
porate perceptions of future fuel availability. State-of-
the-art fuel cell systems demonstrate the potential for
long-term viability: They could realize high electro-
chemical energy conversion efficiencies and very low
local exhaust emissions, depending upon the type of
fuel chosen and the associated reformation process to
produce hydrogen. However, the presence of sulfur in
gasoline could pose a significant problem PEM poi-
soning. Owing to its high potential for reducing fuel
consumption, this emerging technology is receiving
substantial R&D funding. However, most researchers
and automotive manufacturers believe that successful
commercial application of fuel cells for passenger ve-
hicles is at least 10 to 15 years away. Further informa-
tion on fuel cells is provided in a separate section later
in this chapter.
With the exception of fuel cells and series hybrids, the tech-
nologies reviewed above are all currently in the production,
product planning, or continued development stage, or are
planned for introduction in Europe or Japan. The feasibility of
production is therefore well known, as are the estimated pro-
duction costs. However, given constraints on price imposed
by competitive pressures in the U.S. market, only certain tech-
nologies are considered practical or cost effective.
As noted earlier, the exhaust emission standards in the
United States (Tier 2 and SULEV) make the introduction
of some high-fuel-economy technologies, such as lean-
burn, direct-injection gasoline or high-speed DI diesel en-
gines, uncertain. For these technologies to be viable, low-
sulfur fuel must be available and particulate traps and NOx
emissions controls (lean NOx catalyst, NOx trap, SCR)
must be developed. Therefore, current powertrain strate-
gies for gasoline-powered engines use mainly stoichio-
metric air/fuel mixtures, for which three-way-catalyst
aftertreatment is effective enough to meet future emission
standards.
ESTIMATING POTENTIAL FUEL ECONOMY GAINS
AND COSTS
To predict the costs associated with achieving improve-
ments in fuel consumption, it is necessary to assess applica-
tions of the committee's list of technologies in production
vehicles of different types. The committee estimated the
incremental fuel consumption benefits and the incremental
costs of technologies that may be applicable to actual
vehicles of different classes and intended uses. The commit-
tee has hypothesized three successively more aggressive (and
costly) product development paths for each of 10 vehicle
classes to show how economic and regulatory conditions
may affect fuel economy:
Path 1. This path assumes likely market-responsive
or competition-driven advances in fuel economy us-
ing production-intent technology that may be pos-
sible under current economic (fuel price) and regula-
tory (CAFE, Tier 2, SULEV) conditions and could
be introduced within the next 10 years. It holds ve-
hicle performance constant and assumes a 5 percent
increase in vehicle weight associated with safety-
enhancing features.
Path 2. This path assumes more aggressive advances
in fuel economy that employ more costly production-
intent technologies but that are technically feasible for
introduction within the next 10 years if economic and/
or regulatory conditions justify their use. It also main-
tains constant vehicle performance and assumes a 5
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
percent increase in vehicle weight associated with
safety-enhancing features.
Path 3. This path assumes even greater fuel economy
gains, which would necessitate the introduction of
emerging technologies that have the potential for
substantial market penetration within 10 to 15 years.
These emerging technologies require further develop-
ment in critical aspects of the total system prior to com-
mercial introduction. However, their thermodynamic,
mechanical, electrical, and control features are consid-
ered fundamentally sound. High-speed, direct-injec-
tion diesel engines, for instance, are achieving sig-
nificant market penetration in Europe. However, strict
exhaust emission standards in the United States neces-
sitate significant efforts to develop combustion or ex-
haust aftertreatment systems before these engines can
be considered for broad introduction.
.
For each product development path, the committee esti-
mated the feasibility, potential incremental fuel consumption
improvement, and incremental cost for 10 vehicle classes:
· Passenger cars: subcompact, compact, midsize, and
large;
· Sport utility vehicles: small, midsize, and large; and
· Other light trucks: small pickup, large pickup, and
. .
mlmvan.
The three paths were estimated to represent vehicle devel-
opment steps that would offer increasing levels of fuel
economy gain (as incremental relative reductions in fuel con-
sumption) at incrementally increasing cost. The committee
has applied its engineering judgment in reducing the other-
wise nearly infinite variations in vehicle design and technol-
ogy that would be available to some characteristic examples.
The approach presented here is intended to estimate the
potential costs and fuel economy gains that are considered
technically feasible but whose costs may or may not be
recoverable, depending on external factors such as market
competition, consumer demand, or government regulations.
The committee assembled cost data through meetings and
interviews with representatives of automotive manufactur-
ers and component and subsystem suppliers and through
published references. Cost estimates provided by component
manufacturers were multiplied by a factor of 1.4 to approxi-
mate the retail price equivalent (RPE) costs for vehicle manu-
facturers to account for other systems integration, overhead,
marketing, profit, and warranty issues (EEA, 2001~.
Experience with market competition has shown that the
pricing of products can vary significantly, especially when
the product is first introduced. Furthermore, marketing strat-
egies and customer demand can greatly influence the RPE
cost passed along to customers. Retail prices vary greatly,
especially for components required to meet regulatory stan-
dards (such as catalytic converters, air bags, or seat belts).
41
The baseline fuel economies for these evolutionary cases
are the lab results (uncorrected for on-road experience) on
the 55/45 combined cycle for MY 1999 for each vehicle class
(EPA, 2001a). Both the average fuel economy (in mpg) and
the initial fuel consumption (in gallons/100 miles) are shown
in Tables 3-1 through 3-3. The incremental improvements,
however, were calculated as percentage reductions in fuel
consumption (gallons per 100 miles). (The two measures
should not be confused; a 20 percent decrease in fuel
consumption, for example, from 5 gallons per 100 miles to
4 gallons per 100 miles, represents a 25 percent increase in
fuel economy, from 20 mpg to 25 mpg.) The technology
baseline for each vehicle class was set according to whether
the majority of vehicles employed a given technology. Thus,
all cars (but not trucks) are assumed to have four valves per
cylinder and overhead camshafts even though a substantial
number sold in the United States still have two valves, espe-
cially large cars.
The results of this technology assessment are summarized
for passenger cars in Table 3-1, for SUVs and minivans in
Table 3-2, and for pickup trucks in Table 3-3. The distinc-
tion between "production-intent" and "emerging" technolo-
gies for engines, transmissions, and vehicles is maintained.
For each technology considered, the tables give an esti-
mated range for incremental reductions in fuel consumption
(calculated in gallons per 100 miles). The ranges in fuel con-
gumption improvement represent real-world variations that
may result from many (sometimes competing) factors, in-
cluding the baseline state of the engine, transmission, or ve-
hicle; effectiveness in implementation; trade-offs associated
with exhaust emissions, drivability, or corporate standards;
trade-offs between price and performance; differences be-
tween new system design, on the one hand, and carryover or
product improvement on the other; and other calibration or
consumer acceptance attributes such as noise and vibration.
Similarly, the ranges of incremental cost in these tables
represent variations to be expected depending on a number
of conditions, including the difference between product
improvement cycles and new component design; variations
in fixed and variable costs, depending on manufacturer-
specific conditions; commonality of components or sub-
systems across vehicle lines; and evolutionary cost reduc-
tions. In addition, since many of the cost figures were
supplied by component and subsystem suppliers, a factor of
1.4 was applied to the supplied cost to arrive at the RPE to
the consumer.
The analysis presented here is based on the average fuel
consumption improvement and cost of each incremental
technology, as shown in Tables 3-1, 3-2, and 3-3. For each
vehicle class, the average fuel consumption improvement
for the first technology selected is multiplied by the baseline
fuel consumption (adjusted for the additional weight for
safety improvements). This is then multiplied by the average
improvement of the next technology, etc. Costs are simply
added, starting at zero. Figures 3-4 to 3-13 show the incre-
52
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
TABLE 3-5 Published Data for Some Hybnd Vehicles
Power Engine Engine Motor Trans-
Weight Plant Size Power Battery Peak mission CAFE 0-60 Data
Type Status (lb) Type (L) (hp.) Type (kW) Type (mpg)a (see) Source
Toyota Priusb Gasoline
hybrid Com. 2,765 SI I-4 1.5 70 NiMH 33 CVT 58 12.1 c
Honda Insight Gasoline
hybrid Com. 1,856 SII-3 1.0 67 NiMH 10 M5 76 10.6 die
Ford Prodigy Gasoline
hybrid Prot. 2,387 CIDI I-4 1.2 74 NiMH 16 AS 70 12.0 d' f
DC ESX3 Gasoline
hybrid Prot. 2,250 CIDI I-3 1.5 74 Li-ion 15 EMAT-6 72 11.0 ~ g
GM Precept Gasoline
hybrid Prot. 2,590 CIDI I-3 1.3 59 NiMH 35 A4 80 11.5 d, h
NOTE: SI, spark ignition; CIDI, compression ignition, direct injection; CVT, continuously variable transmission; M, manual; A, automatic; EMAT, electro-
mechanical automatic transmission. SOURCE: An (2001).
a CAFE fuel economy represents combined 45/55 highway/city fuel economy and is based on an unadjusted figure.
b U.s. version.
c EV News, 2000, June, p. 8.
MARC (2000a).
e automotive Engineering, 1999, October, p. 55.
f The starter/generator rated 3 kW continuous, 8 kW for 3 minutes, and 35 kW for 3 seconds. We assume 16 kW for a 12-s 0-60 acceleration.
g automotive Engineering, 2000, May, p. 32.
h Precept press release; the front motor is 25 kW and the rear motor is 10 kW, so the total motor peak is 35 kW.
90 -
80 -
70 -
60 -
50 -
c~
40-
30 -
20 -
10-
o- ~
~ l
Hi_
_,
to Do
0 ~
.
38
: 1
u
x
I
C'
.> an.
46
Ad_
~8
>
,~ —
z ~
m
26
~7~
>
o
m
26
I_
~8
>
C0'-
Z ~n
m
26
US Honda Ford DC GM
Prius Insight Prodigy ESX3 Precept
FIGURE 3-15 Breakdown of fuel economy improvements by technology combination. SOURCE: An (2001~.
Hybrid
optimization
Engine
. . .
Downsizing
Load
reduction
Dieseli-
zation
[1 Baseline
Com. Veh.
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
cost, but it will achieve fuel economy similar to its V6, two-
wheel-drive equivalent.
Advanced HEVs cost much more than more conventional
vehicles. In addition, overall system efficiencies must con-
tinue to improve, especially energy conversion, power trans-
fer, electrochemical battery storage, and power output from
the motors. These developments will allow greater overall
fuel efficiency and system trade-offs that would result in
reduced battery and motor sizes, extended electric-only pro-
pulsion range, improved power density, and reduced vehicle
weight.
During the early introduction of these technologies, sev-
eral obstacles must be addressed. First, warranty periods
must be defined and, hopefully, extended with time. Second,
the rate at which battery power systems can accept energy
generated during a hard regenerative braking event must
be improved. Finally, the potential safety consequences
of a depleted battery (loss of acceleration power) must be
clarified.
FUEL CELLS
The emerging technology of fuel cells is also receiving
increasing attention and R&D funding on the basis of its
potential use in passenger vehicles. A few concept vehicles
are now in operation, and a few commercial vehicles may
appear in niche markets in the next few years. Figure 3-16
schematically represents their principle of operation, using
Electrons I)
Porous Electrodes
FIGURE 3-16 Working principles of a PEM fuel cell.
53
hydrogen as a fuel. Hydrogen enters the fuel cell through the
porous anode. A platinum catalyst, applied to the anode,
strips the electrons from the hydrogen, producing a positive
hydrogen ion (a proton). The electrons pass through the load
to the cathode as an electric current. The protons traverse the
electrolyte and proceed to the porous cathode. Ultimately,
through the application of a catalyst, the protons join with
oxygen (from air) and the electrons from the power source to
form water.
Different types of fuel cells employ different materials.
According to Ashley (2001), "the proton exchange mem-
brane (PEM) variety has emerged as the clear favorite for
automotive use." Another type, the solid oxide fuel cell
(SOFC), considered by Ashley and others as a less likely
alternative, is represented by the alkaline air cell. The big-
gest difference between the SOFC and PEM technologies is
their operating temperatures. While PEM cells run at 80°C,
SOFC units run at 700° to 1000°C.
If hydrogen is used as the fuel, no atmospheric pollutants
are produced during this portion of the energy cycle of fuel
cells, since water is the only by-product. No hydrocarbons,
carbon monoxide, NOX, or particulates are produced. It is
important to note that hydrogen could also be used as the
fuel for an internal combustion engine. Louis (2001) quotes
BMW as stating that a "spark ignition engine running on
hydrogen is only slightly less efficient than a direct hydro-
gen fuel cell." The internal combustion engine will produce
a certain level of NOX during combustion. However, due to
No
Load
.
Hydrogen ~{D Air in
Nitrogen
and Water
Vapor Out
54
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
the very lean flammability limit of hydrogen, the NOX con-
centration will be much lower than when using normal hy-
drocarbon fuels.
Energy is consumed, however, and exhaust emissions
are likely to be generated in producing, transporting, and
storing the hydrogen. The energy efficiency of a fuel cell
cycle, "from well to wheels," includes energy losses and
emissions from all of the steps of production, refining, and
distribution of the fuel (see Attachment 4A). Since hydro-
gen is not naturally available, as are conventional fuels, it
must be extracted from other hydrogen-containing com-
pounds such as hydrocarbons or water. Unless the extrac-
tion is performed onboard, the hydrogen must be trans-
ported from the extraction point to the user. Hydrogen
distribution systems are beyond the scope of this report,
but this section evaluates two fuel cell systems that use a
fuel reformer to generate hydrogen onboard the vehicle
from either methanol (which can be produced from natural
gas or biomass) or gasoline.
Onboard reformers have several common difficulties that
must be overcome for commercial acceptance. They typi-
cally operate significantly above room temperature, with
energy conversion efficiencies of 75 to 80 percent. The hy-
drogen is removed from the fuel by either catalysis or com-
bustion. In addition, optimal operation occurs at process
pressures above atmospheric. Furthermore, the response time
and transient power requirements for vehicle application
necessitate some form of onboard storage of hydrogen. For
350
300
250
200
150
100
50
Efficiency
25% Load
Efficiency ~ full
Load
Power Density
Specific Power
FIGURE 3-17 State of the art and future targets for fuel cell development.
commercial success in passenger vehicles, the volume and
weight parameters of the fuel reformer, fuel cell, and electric
drive must be relatively competitive with current power
trains and fuel tanks.
Using methanol as the liquid fuel offers the advantage of
sulfur-free conversion (normal gasoline contains sulfur,
which poisons fuel-cell stacks). Methanol has its own prob-
lems, however. It is toxic if ingested, highly corrosive,
soluble in water (thereby posing a potential threat to under-
ground water supplies), and currently relatively expensive
to produce, compared with gasoline. However, proponents
of methanol cite the toxicity of existing hydrocarbon fuels
(gasoline or diesel fuel) and point out the low evaporative
emissions, the absence of sulfur, and the potential for pro-
duction from renewable sources.
All major automotive manufacturers are actively pursu-
ing fuel cell systems using an onboard reformer. The com-
mittee believes, however, that advanced internal combustion
engine-powered vehicles, including HEVs, will be over-
whelmingly dominant in the vehicle market for the next 10
to 15 years. This conclusion is supported by the recently
released study by Weiss et al. (2000~.
Figure 3-17 shows the state of the art in fuel cell systems
and the targets set by the Department of Energy for longer-
term development. The figure shows that significant devel-
opment advances are necessary to allow the fuel cell to be-
come competitive with the internal combustion engine as a
source of power. It is also important to note that the internal
~ .` ~
300
250
120
34 40 44 ~~ 33 35
325 325
250
1
120
250
125
45
150
150
1 1 ~ 1
Cost
~ c'
0 ~
Resp. Time
(1 0%-90%)
Current
Status
HI Project Goal
[:~:1 Long-Term
Goal
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
0.50 -
~_ it. ~ ~ ~
..s,.~ is.,., *,
a '-
~ ,:'
a....
0.45 - if
~ l;
~1~ 0 40 -
Oh
0.35 -
~ PmaX= 2.5 bar
m" Pmax=2.obar
;;m Pmax=15 bar
Pmax= 1.0 bar
,.~..,
a,...
A.
>..N
.. * ~
S ~
\.p ~
_ b I ~ I
0.30- ~ 1 ~ 1 ~ 1 ~ 1 ~ 1
0 10 20 30 40 50
Effective Power [kW]
FIGURE 3-18 Typical fuel cell efficiency.
55
DOE (Department of Energy). 2000. Fuel Cell and CIDI Engine R&D.
Solicitation #DE-RP04-OlAL67057, November.
Ecker, H.-J. 2000. Downsizing of Diesel Engines: 3-Cylinder/4-Cylinder.
SAE Paper 2000-01-0990. Warrendale, Pa: SAE.
EEA (Energy and Environmental Analysis, Inc.). 2001. Technology and
Cost of Future Fuel Economy Improvements for Light-Duty Vehicles.
Final Report. Prepared for the committee and available in the National
Academies' public access file for the committee's study.
EPA (Environmental Protection Agency). 1999. Regulatory Impact Analy-
sis Control of Air Pollution from New Motor Vehicles: Tier 2 Motor
Vehicle Emissions Standards and Gasoline Sulfur Control Require-
ments. EPA 420-R99-023. Washington, D.C., December.
EPA. 2000a. Light-Duty Automotive Technology and Fuel Economy
Trends 1975 Through 2000. EPA 420-R00-008. Washington, D.C.,
December.
EPA. 2000b. Regulatory Impact Analysis Heavy-Duty Engine and Ve-
hicle Standards and Highway Diesel Fuel Sulfur Control Requirements,
December.
Leone, T.G., E.J. Christenson, and R.A. Stein. 1996. Comparison of Vari-
able Camshaft Timing Strategies at Part Load. SAE Paper 960584.
Warrendale, Pa.: SAE.
Louis, J. 2001. SAE 2001-01-01343. Warrendale, Pa.: SAE.
NRC (National Research Council). 1992. Automotive Fuel Economy: How
Far Should We Go? Washington, D.C.: National Academy Press.
NRC. 2000a. Review of Research Program of the Partnership for a New
Generation of Vehicles: Sixth Report. Washington, D.C.: National
combustion engine will continue to advance as the fuel cell Academy Press.
is being developed.
An additional issue that must be addressed is the reduced
efficiency at higher loads, as shown in Figure 3-18. Substan-
tial development will be required to overcome this character-
istic and other challenges associated with power density, spe-
cific power, production cost, and system response time, before
fuel cells can be successfully commercialized in an HEV.
REFERENCES
An, F. 2001. Evaluating Commercial and Prototype HEV's. SAE 2001-01-
0951. Warrendale, Pa.: Society of Automotive Engineers (SAE).
Ashley, S. 2001. "Fuel Cells Start to Look Real." Automotive Engineering
International. March.
CARE (California Air Resources Board). 1998. Particulate Emissions from
Diesel-Fueled Engines As a Toxic Air Contaminant, November 3.
Available online at
Attachment 3A
A Technical Evaluation of Two Weight- and Engineering-Based
Fuel-Efficiency Parameters for Cars and Light Trucks
Measuring the fuel economy of vehicles in miles per gal-
lon (mpg) alone does not provide sufficient information to
evaluate a vehicle's efficiency in performing its intended
function. A better way to measure the energy efficiency of
vehicles is needed, one that has a sound engineering basis.
This attachment presents two weight-based parameters as
examples of approaches that take the intended use of a ve-
hicle into consideration. One is based on a vehicle's curb
weight and the other includes its payload (passenger plus
cargo). Because of the short time frame for the committee' s
study, an analysis sufficiently detailed to draw conclusions
as to the value of these or other parameters was not possible.
MILES PER GALLON VERSUS GALLONS PER MILE AND HOW TO
MEASURE
The physics of vehicle design can form the basis for pa-
rameters that more accurately represent system energy effi-
ciencies and could be used by EPA in fuel economy testing.
Mpg is not by itself a sufficient parameter to measure effi-
ciency, since it is inherently higher for smaller vehicles and
lower for larger vehicles, which can carry more passengers
and a greater cargo load.
Although CAFE standards currently characterize vehicles
by miles driven per gallon of fuel consumed, the inverse,
gallons per mile, would be more advantageous for several
reasons. As shown in Figure 3A-1, gallons per mile mea-
sures fuel consumption and thus relates directly to the goal
of decreasing the gallons consumed. Note that the curve is
relatively flat beyond 30 or 35 mpg because fuel savings
become increasingly smaller as mpg increases. Also, the use
of fuel consumption (gallons per mile) has analytical advan-
tages, addressed in this attachment. To aid and clarify the
analysis and make the numbers easier to comprehend, the
term to be used is gallons per 100 miles (gal/100 miles). A
vehicle getting 25 mpg uses 4 gal/100 miles.
For reproducibility reasons, fuel consumption measure-
ments are made on a chassis dynamometer. The driving
56
wheels are placed on the dynamometer rollers; other wheels
do not rotate. Thus rolling resistance, as with aerodynamic
drag, must be accounted for mathematically. Vehicle coast-
down times are experimentally determined (a measure of
aerodynamic drag); an auxiliary power unit (APU) ensures
that dynamometer coast-down times are in reasonable agree-
ment with road-tested coast-down times. Test reproducibil-
ity is in the few percent range. The driver follows two differ-
ent specified cycles, city and highway, which were deduced
from traffic measurements made some 30 to 40 years ago. A
change in the test cycle is not a minor item much engineer-
ing know-how is based on the present cycle, which is also
used for exhaust emissions measurements.
WEIGHT SPECIFIC FUEL CONSUMPTION
Figure 5A-4 (in Chapter 5) plots gal/100 miles versus
vehicle weight for MY 2000 vehicles. The vertical scatter
along a line of constant weight reflects the fact that vehicles
A_
in
~ 12-
o
o
10 - \
8-
6-
4-
2-
O-
|—Series1 |
\
\
10 15 20 25 30 35 40 45 50 55 60
Fuel economy (miles per gallon)
FIGURE 3A- 1 Dependence of fuel consumption on fuel economy.
SOURCE: NRC (2000~.
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
of the same weight may differ in the efficiency of their drive
trains or rolling resistance or aerodynamic drag (and thus in
the number of gallons used to travel a given distance). While
gal/100 miles is a straightforward parameter for measuring
fuel consumption, it does not reflect the load-carrying ca-
pacity of the vehicle. Smaller cars, with lower fuel consump-
tion, are designed to carry smaller loads, and larger cars and
trucks, larger ones.
For engineering analysis purposes, it is convenient to nor-
malize the data in Figure 5A-4, that is, divide the y value
(vertical scale) of each data point by its curb weight in tons.
The resulting new vertical scale is the weight-specific fuel
consumption (WSFC). The units shown in Figure 3A-2 (and
Figure 5A-5) are gal/ton of vehicle weight/100 miles. The
straight horizontal line is a reasonable representation of the
average efficiency of fuel use data for a wide variety of ve-
hicle types and weights. It shows that the efficiency (WSFC)
is approximately the same for this variety of different ve-
hicle types (MY 2000, 33 trucks and 44 cars) and weights.
Note that some vertical scatter is to be expected; all vehicles
having approximately the same weight do not necessarily
have the same dnve-tra~n efficiency.
Figure 3A-3 shows on-the-road data taken by Consumer
Reports (Apnl 2001~. Their measurements were based on a
realistic mixture of expressway, country-road, and city dnv-
4 -
~n
. _
~ 3 -
o
A
s
.O
c'
> 2 -
a
o
t
o
is
CD 1 -
IL
u'
O-
2000 2500 3000 3500
57
AUTO
LTR&SUV
HYBRID
8
0 2000 4000 6000 8000
vehicle weight (pounds)
FIGURE 3A-3 Fleet fuel economy. Based on information from
Consumer Reports (April 2001~.
ing. Again, the efficiency of fuel use for their on-the-road
tests is reasonably represented by a horizontal straight line.
Figure 3A-3 illustrates the analytical utility of this approach.
The lowest car point, at a little less than 3,000 lb, is a diesel
engine; its WSFC is around 1.8 compared with around 2.7
for the average. The two hybrid points also show lower
WSFC than the average but higher than the diesel.
Figure 3A-4 illustrates possible realistic reductions in
WSFC. EPA has fuel-consumption data for more than a thou-
X
Xx
\v
· {it ~ ~ ~ ~ ~~ ~ In; +;~. ,~.· ; ; ~
Cars Trucks
· ~
~ Premium Small + Comp VAN
· Entry Small X Comp PU
i\ Midsize >K STD PU
O Near Luxury Midsize ~ Small SUV
· Luxury Midsize ~ Comp SUV
O Large ~ Luxury Comp SUV
· Luxury Large ~ Large SUV
- - Average
.
4000 4500 5000 5500
Weight, lb
FIGURE 3A-2 Weight-specific fuel consumption versus weight for all vehicles.
58
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
· AUTOS
~ LT TRUCKS
|A HYBRIDS
co
~ 1.6-
o
or
Q
o
c' 0.8-
CO
O 0.4-
C) O- 1 1 1
an
o
2000 4000 6000
Vehicle weight (pounds)
FIGURE 3A-4 Best-in-class fuel-efficiency analysis of 2000 and
2001 vehicles.
sand MY 2000 and 2001 vehicles. A horizontal line was
drawn on a WSFC (highway) graph for these vehicles such
that 125 vehicles were below this arbitrary line. The results
for the 125 vehicles are shown in Figure 3A-4. The average
WSFC value for all vehicles was 1.7; the average for the 125
vehicles was around 1.4. Since these were production ve-
hicles, it would appear that application of in-production tech-
nologies to the entire fleet could produce significant reduc-
tions in WSFC.
LOAD-SPECIFIC FUEL CONSUMPTION
For a heavy-duty truck designed to carry a large payload,
the most meaningful parameter would be normalized by di-
viding the gallons per mile by the tons of payload, to arrive
at a load-specific fuel consumption (LSFC), that is, gallons
per ton of payload per 100 miles. This number would be
lowest for vehicles with the most efficient powertrain sys-
tem and the least road load requirements (lightest weight,
low accessory loads, low rolling resistance, and low drag)
while moving the largest payload. Similarly, a reasonable
parameter for a fuel-efficient bus would be gallons per pas-
senger-mile.
The parameter LSFC is more difficult to define for light-
duty vehicles than for heavy-duty trucks or buses, because
the payloads are widely different (and harder to define) for
these vehicles. This report calculates a total weight capacity
by multiplying passenger capacity (determined by the num-
ber of seat belts) by an average weight per person (150 lb)
and adding cargo weight capacity, which is the cargo vol-
ume multiplied by an average density (say, 15 lb/ft3~.i For
iThis is an estimated density for cargo space. GM uses about 11 lb/ft3
across a range of vehicles. Further study needs to be done to determune a
representative design density to use.
pickup trucks, the difference between gross vehicle weight
(GVW) and curb weight was used to determine payload. The
weight of the passengers and cargo could be added to the
vehicle's weight, and the sum used in the EPA fuel economy
test to determine engine loading for the test cycle. Alterna-
tively, the present fuel economy data could be used with the
above average passenger plus cargo weight. The fuel con-
sumption on the city and highway cycle would be measured
and expressed as gallons/ton (passenger plus cargo weight)/
100 miles.
Figure 3A-5 plots fuel economy against payload in tons
for heavy-duty and light-duty vehicles. Lines of constant
mpg-tons are also shown, with larger numbers representing
more efficient transport of payloads. LSFC, the inverse of
mpg-tons, is also shown on the lines, with lower numbers
representing lower normalized vehicle fuel consumption.
The point representing the PNGV goal is also shown.
Fuel economy measures based on this parameter would
drive engineers to maximize the efficiency with which ve-
hicles carry passengers and cargo while minimizing struc-
tural weight. This new fuel consumption parameter has the
potential to be a better parameter to compare different types
and sizes of vehicles.
Figure 3A-6 graphs the same light-duty vehicles' fuel
consumption (in gal/100 miles) as a function of the payloads
in Figure 3A-8. This figure shows the large difference in fuel
consumption between cars (2.5 to 4 gal/100 mi) and trucks
(3.5 to 5.5 gal/100 ml); the CAFE standards for both types of
vehicles are included for reference. Figure 3A-7 shows LSFC
(in gallons per ton of passengers plus cargo) for the same
vehicles. LSFC appears to normalize fuel consumption,
bridging all types of vehicles. Both types of vehicle re-
gardless of size and weight are represented above and be-
low the average line. This finding suggests that LSFC is a
good engineering parameter for both cars and trucks.
COMPARING THE TWO WEIGHT BASED PARAMETERS
The WSFC essentially normalizes the fuel consumed per
100 miles to take out the strong dependence on vehicle
weight. Different weight vehicles can be compared more
equitably. Lower WSFC parameters indicate lower road load
requirements and/or higher powertrain efficiencies with
lower accessory loads.
Figure 3A-9 shows fuel economy versus vehicle curb
weight for the 87 light-duty vehicles. Constant efficiency
lines (in mpg x tons of vehicle weight) are shown along
with WFSC. Figure 3A-10 shows fuel economy versus ve-
hicle payload for the 87 vehicles. The constant efficiency
lines in mpg x tons of payload are shown along with LSFC.
The utility of this plot is that it shows the interrelationship
of fuel economy, payload, and LSFC in gallons/payload
tons/100 miles. LSFC and WSFC show similar utility in
determining whether certain types of vehicles are either
above or below the average lines in Figures 3A-2 and
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS
2.5
, us -
90 -
80 -
70 -
60 -
Q
-
° 50
o
-
IL
40
30
20
10 -
o
8.3 1.1 0.5 0.25
1;
. . it,,
.~
LSFC, gallons/Payload Ton -100 miles
-
\
/A/ / //> f ~ f -7` \ Heavy-Duty Vehicles ~
~\ ~ ~ I_
0 5 10 15
FIGURE 3A-5 LSFC versus payload for a variety of vehicles.
10 -
9
8 -
7 -
~n
._ ~
~ V -
o
o
In
o
Ct
-
IL
5 -
4 -
3 -
2 -
1 -
~[11
20 25 30 35 40
Payload, tons
Cars
Trucks
y = 0.O2O1x + 0.6132
R =0.804
>A _
C1 Premium Small
· Entry Small
Midsize
O Near Luxury Midsize
· Luxury Midsize
0 Large
· Luxury Large
- CAFE Standard for Cars
+ Comp VAN
X Comp PU
OK STD PU
Small SUV
:: Comp SUV
88 Luxury Comp SUV
· Large SUV
CAFE Standard for Trucks
O -
2000 2500 3000 3500 4000 4500 5000 5500
FIGURE 3A-6 Fuel consumption versus payload.
Payload/lb
59
60
FIGURE 3A-7 Payload versus LSFC.
2400 -
2200 -
2000 -
1800 -
1600 -
1400 -
g
10 -
9
tn
~ 8 -
Q
tn
~ 7 -
tn
tn
ct
~ v -
o
tn
O 5
1
tn
~ 4 -
o
0
~n
o
= ~
ct '
-
IL 1 -
~n
O -
[]
_ _ _ _
[1
[1
·0—
~0 0
~\ ~ ~] 0~0
_~_ _ _____
_
~X~
~d X
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
.
.
-
+
~K OC~
o
+
·
Cars
Trucks
Premium Small
· Entry Small
Midsize
0 Near Luxury Midsize
· Luxury Midsize
0 Large
· Luxury Large
- - Average
l
X
~v
m
::
Comp VAN
Comp PU
STD PU
Small SUV
Comp SUV
83 Luxury Comp SUV
· Large SUV
.
2000 2500 3000 3500 4000 4500 5000 5500
Weight, lb
X
X
1200 -
1 000 -
800 -
600 -
400 -
200 -
.
_~~
/\ ~ ~ ~ y 51.59x + 821.64
~ O ~ R2= 0.4757
~ {~ ~ S ° t
In order of increasinq averaqe weiqht
o 1 , -
Ct Ct N ~ ~ N N ~ ~ ~ ~ ~ ~ >
~ ~ ~ ~ Q ~ ~ ~ > ~ Q ~ Q
~ E ~ ~ ~ 2~ ~ x O O
x x ,
' x
FIGURE 3A-8 Payload for a variety of vehicles.
TECHNOLOGIES FOR IMPROVING THE FUEL ECONOMY OF PASSENGER CARS AND LIGHT-DUTY TRUCKS 61
3.8 3.1 2.6 2.3 2.1 1.9 WSFC, gallons/Ton of Vehicle Weight -100 miles
50- \ \ \ \ \
45 -
40 -
35 -
30-
-
° 25-
o
~ 20-
IL
15 -
10 -
5 -
O -
.
1000 1500 2000
"''" ''''' :
\ \ art\ at' '' ''''
\ \ \~\4 ~
i.
. ...
2500 3000
_ _
Oh oh
~ .=
. . .
3500 4000 4500 5000 5500 6000 6500 7000
Weight, lb
>=N N N0) ~> ~
FoO ~ ~~ oX On
O~< X X O
Z
cn
FIGURE 3A-9 Fuel economy as a function of average WSFC for different classes of vehicles.
3A-8. Using LSFC, however, would encourage manufac-
turers to consider all aspects of vehicle design, including
materials, accessory power consumption, body design, and
engine and transmission efficiency. The use of the LSFC
number will show high-performance, heavy, two-seat
sports cars without much cargo space and large luxury cars
to be on the high side compared with vehicles designed to
be fuel- and payload-efficient.
WSFC does not account for the load-carrying capacity of
Cars
C1 Premium Small
· Entry Small
/\ Midsize
0 Near Luxury Midsize
· Luxury Midsize
0 Large
· Luxury Large
Trucks
+ Comp VAN
XComp PU
KSTD PU
Small SUV
:: Comp SUV
8~9 Luxury Comp SUV
· Large SUV
certain vehicles such as pickups, vans, and SUVs. The vans
and large SUVs in Figure 3A-8 are shown below the average
fit line and below the average WSFC line in Figure 3A-2,
indicating they have highly efficient powertrain technologies
and low road load/accessory load requirements. Pickups (PUB)
are above the average WSFC line in Figure 3A-2, showing
that it is difficult to design a truck that has low aerodynamic
drag and is fuel efficient. When the fuel consumptions of these
vehicles (vans, pickups, and SUVs) are normalized to pay-
62
5O1
451
401
35
~ 30
° 25
8
20
IL
15
10 1
5
O
Payload, lb
~ if:! in— `` E E !1 t; E E
FIGURE 3A-lO Fuel economy versus average payload for different classes of vehicles.
load, they are below the average LSFC line in Figure 3A-8, REFERENCE
indicating they are well designed for their intended use.
CONCLUSION
Both the WSFC and LSFC parameters have potential util-
ity as fuel-efficiency parameters for vehicles, but their appli-
cability requires additional study.
EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS
12.5 8.3 6.3 5.0 4.2
\ \ \ \: ::::: :
\ \ \ ~
\ \ : :~: :\:
\ hi: .:N : ~ ,,
\ ~ he'd"''\' '\''
\ ~ ~ ~
LSFC, gallons/Payload Ton -100 miles
\ \ :~: : ~ :: \ : :
\ \2]~2~ ~
At.
If 0W
\ .
\ .
\ .....
.....
lo
.—.
..... .
·
. ..... .
..... . . . ~
' . . ~
..... . . . . . .
..... . . . . . .
..... . . . . . .
..... . . . . . .
: ::::: : : :: : : :
. ..... . , ..
0 400 800 1200 1600 2000 2400
24 ~~O ~ a -_
:z
- ~16 mpg x O.STon
Cars
C1 Premium Small
· Entry Small
~ Midsize
0 Near Luxury Midsize
· Luxury Midsize
0 Large
· Luxury Large
Trucks
+ Comp VAN
x Comp PU
>KSTD PU
Install SUV
:: Comp SUV
88 Luxury Comp SUV
· Large SUV
NRC (National Research Council). 2000. Automotive Fuel Economy: How
Far Should We Go? Washington, D.C.: National Academy Press. p. 156.
