Tires and Passenger Vehicle Fuel Economy: Informing Consumers, Improving Performance -- Special Report 286 (2006)

In every important respect, the quality and performance of today’s passenger tires are superior to those of their predecessors. Tires wear longer, are more resistant to damage, handle and track better, and are easier to maintain. Each generation of tire engineers has sought to balance these and other performance characteristics, commensurate with technology cost and capabilities, government regulations, consumer demands, and operational requirements.

In requesting this study, Congress did not give specific reasons for its interest in tire energy performance. However, it did ask for estimates of the fuel savings associated with low-rolling-resistance tires. Accordingly, the committee construed its charge to focus on the contribution of tires to passenger vehicle fuel consumption, as opposed to all energy flows during a tire’s life cycle, from the energy used in raw materials and manufacturing processes to recycling and disposal. While a full accounting of such life-cycle effects is relevant for policy making, it would have exceeded the scope and capabilities of this study.

The chapter begins with a review of the history of interest in vehicle fuel economy and the effect of tires on fuel consumption. Rolling resistance, which is the main source of the tire’s influence on fuel consumption, is then explained. Over the past 25 years, several data sets containing measurements of the rolling resistance characteristics of new tires have been made available to the public. These data sets are examined. Although they are limited in coverage, they offer insights into changes in rolling resistance over time and the implications for passenger vehicle fuel economy.

Fuel economy is typically expressed as the average number of miles a vehicle travels per gallon of motor fuel, usually as miles per gallon (mpg). The interest of both consumers and government in fuel economy was galvanized during the mid-1970s in response to escalating fuel prices prompted by the oil embargo of the Organization of Petroleum Exporting Countries. At that time, new cars sold in the United States averaged less than 16 mpg. As gasoline prices jumped by more than 25 percent within months, motorists and policy makers focused their attention on energy conservation for the first time since World War II. During the decade that followed—which included further jumps in gasoline and diesel fuel prices—the average fuel economy of new vehicles grew by more than 50 percent (NRC 1992, 14). During this period some policy makers also began to focus on the role of motor fuel in the atmospheric buildup of carbon dioxide and other greenhouse gases. The buildup threatened climate change and provided further impetus for improvements in fuel economy (TRB 1997).

A number of policies aimed at energy conservation were pursued starting in the mid-1970s. Congress passed the national 55-mph speed limit in 1974. A year later, it instructed the U.S. Environmental Protection Agency (EPA) to require the posting of fuel economy labels (window stickers) on all new vehicles for sale. The U.S. Department of Energy was charged with developing and publicizing an annual fuel economy mileage guide. The federal “gas guzzler” excise tax, which raised the price of automobiles with low fuel economy, was introduced in 1979. Perhaps the most significant program originating from that period was the corporate average fuel economy (CAFE) program.¹ For the first time, Congress established fuel economy standards for passenger cars and light trucks. The program, administered by the National Highway Traffic Safety Administration (NHTSA), mandated a sales-weighted average fuel economy for different vehicle categories produced by all automobile manufacturers. Each vehicle’s rating would be determined by EPA’s city

and highway driving tests developed originally for emissions testing and certification.²

There are various ways to increase vehicle fuel economy. Among them are reducing the loads that must be overcome by the vehicle and increasing the efficiency of its engine, its transmission, and other components that generate and transfer power to the axles. Since the 1970s, the emphasis given to specific means has fluctuated in response to regulation, market forces, and technology cost and capabilities. At first, automobile manufacturers focused on reducing vehicle mass, most commonly by moving to smaller vehicles constructed of lighter materials (NRC 1992). By the 1980s, the emphasis shifted to increasing engine and transmission efficiency and reducing other vehicle loads such as aerodynamic drag and the power demanded by accessories (NRC 1992). By the end of the 1980s, however, fuel economy gains in passenger cars and light trucks had flattened out. At the same time, gasoline prices had fallen back and public demand for fuel economy waned (NRC 1992, 17).

While modest additional improvements in fuel economy were made during the 1990s, the average fuel economy of the passenger vehicle fleet had already peaked. As larger and more powerful vehicles came back in demand, the modest fuel economy improvements that did occur were achieved by changes in vehicle features not affecting vehicle size or interior space, such as accessories, construction materials, lubricants, and tires. Continuing improvements in engine efficiency were also sought to maintain fuel economy as the market shifted to larger and more powerful vehicles.

Most recently, in a period characterized by higher gasoline prices, mounting concern over national security, and growing consumer interest in fuel economy, NHTSA has set light truck standards to increase at about 0.5 mpg per year from 2005 through 2011. Passenger car standards have not been changed. It is notable, however, that NHTSA and EPA are revising the long-standing means of measuring and calculating vehicle fuel economy, which could eventually affect the implementation of CAFE.

The advent of CAFE and other government policies to promote fuel economy prompted automobile manufacturers and engineers to take a closer look at the many factors influencing vehicle fuel consumption. While explanations of these influences are available elsewhere (Schuring 1980; Ross 1997; NRC 2002; Sovran and Blaser 2003), a general overview is helpful in understanding the contribution of tires to energy consumption.

The amount of fuel consumed by a motor vehicle over a distance is affected by the efficiency of the vehicle in converting the chemical energy in motor fuel into mechanical energy and transmitting it to the axles to drive the wheels. Figure 3-1 depicts the energy flows and sinks for a conventional gasoline-powered midsize passenger car. Most of the energy available in the fuel tank—about two-thirds—is lost in converting heat into mechanical work at the engine, much of it unavoidably. For urban trips consisting of stop-and-go driving, a significant percentage (about 15 to 20 percent) is also lost in standby operations during coasting, braking, and idling in traffic. For urban driving, only 10 to 15 percent of the fuel energy is ultimately transmitted as power to the wheels. Because standby losses are lower during highway driving and because the engine is operating more efficiently, a higher percentage of fuel energy—about 20 percent—makes its way to the wheels. While the specific percentages will vary by vehicle type and trip, the flows shown in Figure 3-1 are generally representative of passenger vehicles today.

For both urban and highway driving, the mechanical energy that does make its way through the driveline to turn the wheels is consumed by three sinks: aerodynamic drag, rolling resistance, and braking. Braking consumes momentum from the vehicle, which must be replenished by acceleration. Because frequent stopping and starting entail repeated braking and acceleration, braking is a major consumer of mechanical energy during urban driving. In contrast, aerodynamic drag consumes relatively more energy during highway driving since this resistive force escalates with vehicle speed.

In comparison, the energy losses from rolling resistance (for a given vehicle and set of tires) are mainly a function of miles traveled. For reasons explained later in this chapter, vehicle speed has a limited effect on

rolling resistance except at the highest speeds reached on occasion during highway driving. As a result, the energy lost per mile because of rolling resistance will be similar for a given vehicle and set of tires over a wide range of urban or highway driving cycles. While the percentage contribution of rolling resistance to total energy consumed per mile depends on the contribution of other sinks, its absolute contribution does not.

In sum, for most conventional motor vehicles in common use, the majority of the energy contained in motor fuel is dissipated as unrecoverable heat from engine combustion and friction in the engine, driveline, axles, and wheel bearings. Some of the energy output from the engine is used during idling and to power vehicle accessories. Only about 12 to 20 percent of the energy originating in the fuel tank is ultimately transmitted through the vehicle’s driveline as mechanical energy to turn the wheels. Rolling resistance consumes about one-third of this mechanical energy output. Rolling resistance, therefore, directly consumes a small portion (4 to 7 percent) of the total energy expended by the vehicle. However, reducing rolling resistance, and thus reducing mechanical energy demand, by a given amount will translate into a larger reduction in total fuel consumption because less fuel energy will need to be sent to the engine in the first place. The effect on total fuel consumption will depend on a number of factors, including the efficiency of the engine and driveline as well as the amount of energy used by accessories.

As explained later in this chapter, for most passenger vehicles, a 10 percent reduction in rolling resistance will lead to a 1 to 2 percent increase in fuel economy and a proportional reduction in fuel consumption. This assumes that other influences on fuel consumption are held constant, especially miles of travel. As a practical matter, total travel by the U.S. passenger vehicle fleet continually increases; it has grown by an average of 1 or 2 percent annually during the past several decades. Accordingly, the time frame over which the change in fuel economy occurs—in the near term or over a longer period—is important in calculating the national fuel savings. A related issue is that improvements in vehicle fuel economy have the secondary effect of increasing vehicle travel. As vehicle fuel economy improves, the per-mile cost of driving is effectively lowered, which may spur some additional driving and fuel consumption. This response, known as the rebound effect, is usually considered in evaluations

of CAFE and other fuel economy programs. After examining the literature, Small and Van Dender (2005) estimate that 2 to 11 percent of the expected fuel savings from a fuel economy improvement is offset by increased driving. While this second-order effect is recognized again later in the report, the calculations of fuel savings do not account for it. For simplicity, it is assumed that miles traveled are unchanged.

Estimates of consumer fuel savings from reductions in rolling resistance are made in Chapter 5. The focus of the remainder of this chapter is on describing the factors causing and influencing rolling resistance as well as the properties of today’s passenger tires with respect to this characteristic.

Short of changing the characteristics of the road surface, there are two main ways to minimize rolling resistance. One is to drive on properly inflated and aligned tires. The other is to use tires that possess low rolling resistance at proper inflation levels. Maintaining proper tire inflation and alignment is important for motor vehicle safety as well as for fuel economy; this is true for all pneumatic tires regardless of their design. This section therefore focuses on designing tires with lower rolling resistance when properly inflated.³

It has long been known that a rolling tire must be supplied energy continuously in order to avoid losing speed. Until the 1970s, however, understanding the causes of tire rolling resistance drew little interest (Schuring 1980). Only a few dozen technical papers had been published on the subject, and no standard methods were in place for measuring tire rolling resistance characteristics (Clark 1983). Rising energy prices during the 1970s prompted more concerted efforts to highlight the causes of rolling resistance and the effects of specific tire construction properties on this characteristic.

With the aid of advances in analytical and experimental capabilities, such as thermography and finite element modeling, tires were examined for a wide range of design, operating, and environmental conditions that could affect rolling resistance. Consideration was given to the effect of tire dimensions, construction types, and materials; load and inflation pressures; wheel alignment; steering and torque inputs; vehicle operating speeds; and ambient temperatures (Clark and Dodge 1978; Schuring 1980).⁴ Even the contributions of roadway surface types and textures were examined (DeRaad 1978; Velinsky and White 1979).

Because of this research, much more is known and documented today about the sources of rolling resistance and their interacting effects.

Pneumatic tires offer a number of advantages related to the highly compliant nature of rubber. The rubber tire interacts with the hard road surface by deforming under load, thereby generating the forces responsible for traction, cornering, acceleration, and braking. It also provides increased cushioning for ride comfort. A disadvantage, however, is that energy is expended as the pneumatic tire repeatedly deforms and recovers during its rotation under the weight of the vehicle.

Most of this energy loss stems from the viscoelastic behavior of rubber materials. Rubber exhibits a combination of viscous and elastic behavior. A purely elastic material is one in which all energy stored in the material during loading is returned when the load is removed and the material quickly recovers its shape. A purely viscous material, on the other hand, stores no strain energy, and all of the energy required to deform the material is simultaneously converted into heat. In the case of a viscoelastic material, some of the energy stored is recovered upon removal of the load, while the rest is converted to heat. The mechanical energy loss associated with each cycle of deformation and recovery is known as hysteresis.⁵

The characteristics affecting hysteresis are a tire’s design and construction and the material types and quantities used.

The beneficial effect of radial-ply constructions in reducing tire rolling resistance is an example of the influence of tire construction on hysteresis. In comparison with the bias-ply tire, the steel-belted radial tire reduced the deformation of the tread in the contact patch. Hence, in addition to affecting tire handling, endurance, and ride comfort, the changeover from bias-ply to radial-ply tires during the 1970s and 1980s reduced tire rolling resistance by an estimated 25 percent without requiring major changes in the polymers used (Schuring 1980, 601).

There are several measures of the geometry of a tire, including its outer diameter, rim diameter, and width. Reducing a tire’s aspect ratio— that is, its section height relative to its section width—should reduce hysteresis if it is accomplished by shortening and stiffening of the sidewalls. The aspect ratio, however, can be altered in other ways—for instance, by changing the tire’s outer diameter, width, rim diameter, or all three dimensions. Moreover, changing tire geometry is difficult without changing other characteristics of the tire that influence hysteresis, such as mass, material types, and construction features. As a result, it can be difficult to know, a priori, how specific changes in tire dimensions will translate to changes in rolling resistance (Schuring 1980; Chang and Shackelton 1983; Schuring and Futamura 1990; Pillai and Fielding-Russell 1991).

Because hysteresis is fundamentally related to the viscoelastic deformation of the rubber used in tire construction, changes in material formulations and quantities affect rolling resistance. While reducing the amount of hysteretic material in any component of the tire might appear to be a straightforward way to reduce rolling resistance, different components must contain different amounts and types of hysteretic material. In particular, the tread contains much of the hysteretic material in the tire. Not only is the tread made of rubber compounds that are designed to improve wet traction, the tread band also contains relatively large quantities of material to prolong wear life. Studies indicate that the tread alone can contribute more than half of hysteretic energy losses in a tire (Chang and Shackelton 1983; Martini 1983; LaClair 2005).

Related to the effect of tread mass and volume on hysteresis is the effect of tread wear on rolling resistance. As tread depth (that is, the depth of grooves in the tread pattern) diminishes with wear, a tire loses about 15 percent of its mass—since the tread band typically accounts for about one-quarter of a tire’s weight. The moderating effect of tread wear on rolling resistance has been examined and quantified to some extent. Martini (1983) compared the tire rolling resistance occurring when the tread was new (100 percent) with that occurring when the tire was buffed to various stages of wear (75, 50, 25, and 0 percent remaining tread). These experiments suggested that rolling resistance declined by 26 percent over the entire wear life. After reviewing many similar experimental studies conducted before 1980, Schuring (1980, 683–684) concluded that rolling resistance declined by an average of about 20 percent over the tread life, dependent on design details.

The tread compound consists of rubbers that contain different polymers, reinforcing fillers, extender oils, antidegradants, and other materials. Their effect on rolling resistance can be significant but complex. Compounding material formulas are developed with many requirements and performance properties in mind. Therefore, these formulas tend to be proprietary, and the rolling resistance effects of different materials and their interactions are difficult to study. The type of rubber used influences rolling resistance; notably, synthetic rubbers tend to exhibit greater rolling resistance than natural rubbers. The reinforcing fillers in the compound, which are essential for abrasion resistance, also affect rolling resistance. Carbon black is the most widely used filler. During the early 1990s, Michelin introduced a silica filler in conjunction with a silane coupling agent as a means of reducing rolling resistance while retaining wet traction characteristics. Although carbon black remains the predominant filler, all major tire companies have reportedly constructed tires containing silica–silane and carbon black in the tread compound. This technology, initially promoted as a breakthrough in the ability to balance rolling resistance with other tire performance properties, is examined in more detail in Chapter 5.

A number of tire operating conditions affect rolling resistance. The most important are load, inflation pressure, and temperature. Tires operated

at the top speeds associated with normal highway driving may exhibit increases in rolling resistance as the frequency of tire deformation increases. However, as speed increases, the tire’s internal temperature rises, offsetting some of the increased rolling resistance. The net effect is that operating speed tends to have a small influence on rolling resistance compared with that of many other operating variables under normal driving conditions (Schuring 1980, 638; Schuring and Futamura 1990, 351; Chang and Shackelton 1983, 19; Hall and Moreland 2001, 530; LaClair 2005, 491). Another nontire operating condition, the road surface, can have an appreciable effect on rolling resistance, as discussed briefly later.

The more a tire at a given pressure is loaded, the more it deforms; hence, hysteresis increases with wheel load. Indeed, the relationship between rolling resistance and sidewall deflection due to load is approximately linear, so increasing the load on a tire results in a near-proportional increase in total rolling resistance. As described later, this linear relationship allows rolling resistance to be expressed as a coefficient with respect to load under normal operating conditions.

Inflation pressure affects tire deformation. Tires with reduced inflation exhibit more sidewall bending and tread shearing. The relationship between rolling resistance and pressure is not linear, but it is consistent enough for rules of thumb to be applied. Schuring (1980) observes that for conventional passenger tires, an increase in inflation pressure from 24 to 29 pounds per square inch (psi) will reduce rolling resistance by 10 percent. For a tire inflated to pressures between 24 and 36 psi, each drop of 1 psi leads to a 1.4 percent increase in its rolling resistance. The response is even greater for pressure changes below 24 psi. Maintenance of tire pressure is therefore important in preventing excessive deformation and hysteresis, as well as in achieving intended wear, traction, handling, and structural performance.

The temperature of a tire is affected by ambient conditions, tire design and materials, running time, and speed. Higher ambient temperatures are associated with reduced rolling resistance because the amount of energy dissipated when the rubber is subjected to repeated deformation declines moderately as temperature rises, which is a commonly observed behavior of viscoelastic materials. Accordingly, the length of time a tire has been running since the last cool-off affects rolling resistance, which declines until the passenger tire has been rolling for about 30 minutes.

At that point an equilibrium temperature is reached and rolling resistance stabilizes.

Researchers have known for some time that rough road surfaces contribute to rolling loss by exacerbating tire deformation. This effect can increase energy losses by 5 to 20 percent (Velinsky and White 1979; DeRaad 1978). Road roughness has two components: macrotexture and microtexture. The first relates mainly to the surface condition on a scale of inches to feet and reflects the presence of cracks, ruts, bumps, and other surface irregularities. Macrotexture can include intentional changes in surface texture, such as surface grooving to improve water runoff. The second component, microtexture, relates to smaller-scale asperities in the road surface that are millimeters or even fractions of a millimeter in size and reflect the coarseness of the surface texture. Tires operated on a rough macrotexture or rough microtexture will deform more and suffer greater energy loss. They will also experience faster tread wear.

The roadway can also contribute to rolling resistance by deflecting or deforming under the weight of the wheel load. How much energy is lost will depend on the rigidity of the roadbed and overlay. Dirt and gravel roads deform the most and give rise to twice as much rolling resistance as harder paved surfaces (DeRaad 1978). However, most driving occurs on paved surfaces, which can vary in rigidity depending on the overlay, base, and subgrade. The most rigid, or nondeformable, pavements tend to be those with a concrete surface layer and reinforced base, followed by an asphalt surface on a concrete base, and an asphalt surface on a compacted gravel or soil base. The rigidity of asphalt overlay depends on the amount and type of asphalt used in relation to aggregate and on environmental conditions such as temperature. A rubber-modified asphalt overlay (often derived from the ground rubber of scrap tires) will deform more under load and thus should create more rolling resistance than harder asphalt pavements.

The fact that rolling resistance relates linearly to wheel load allows it to be expressed as a near-constant coefficient relative to wheel load. The rolling resistance coefficient (RRC), referred to extensively throughout

the remainder of this report, is derived by dividing rolling resistance by wheel load. It is typically measured for new tires—as is the case for many of the data presented in the remainder of this report—but can be measured at any point in a tire’s lifetime. For most passenger tires sold in the United States, the coefficient of the tire measured when it is new falls between 0.007 and 0.014. Hence, for a tire in this range under a load of 1,000 pounds, the rolling resistance is 7 to 14 pounds, resulting in 28 to 56 pounds of total force for the four tires on a vehicle weighing 4,000 pounds, including passenger and cargo load. At 60 mph, a total rolling resistance of 40 pounds consumes about 7 horsepower.

The Society of Automotive Engineers (SAE) has established two standard procedures for measuring tire rolling resistance. Because the procedures, J1269 and J2452, are both laboratory tests, they allow for repeatability and instrumentation accuracy as well as controls for operating conditions and other exogenous influences. They are described in detail in the Appendix. What distinguishes the two test procedures the most is that the first measures rolling resistance at a single speed (50 mph), while the latter measures it over a range of speeds. J1269 was developed to assist tire engineers in quantifying rolling resistance in a consistent way to allow for the more precise balancing of this tire property with other quantifiable properties, such as cornering, traction, and heat generation. J2452 was developed later to provide additional quantification of a tire’s rolling resistance for more precise inputs to the driving cycles used for federal vehicle emissions and fuel economy regulatory compliance. The speed-adjusted measurements generated from J2452 can be entered into simulated driving cycles, such as those used for testing new vehicles for CAFE compliance.

By providing established and commonly accepted methods for measuring tire rolling resistance, the SAE procedures allow reliable comparisons of tires. Of course, neither procedure can take into account all the conditions an individual tire will experience under varied driving and operating conditions over tens of thousands of miles. Variations in road surfaces, inflation pressures, wheel maintenance and alignment, and other conditions will affect rolling resistance in the field. All of these factors— as well as limited correlation of testing equipment—will lead to some discrepancies among individual laboratory measurements and between laboratory results and field experience.

For the most part, the SAE tests are performed only on new tires, and thus they offer little insight into how individual tires experience changing rolling resistance as they are used, wear, and age.⁶ As will be discussed in more detail later, the absence of rolling resistance data for tires at different stages of their use makes it difficult to calculate average rolling resistance and thus to know precisely how one tire’s lifetime energy performance will differ from that of another.

Knowledge of a tire’s RRC allows calculations of its effect on vehicle fuel economy. Such calculations have been the subject of empirical models, laboratory experiments, and road measurements for many years. General approximations, or rules of thumb, of the fuel economy effects of incremental changes in tire rolling resistance have been developed. The most common way to describe this relationship is by relating the percentage change in RRC to the percentage change in fuel economy; for example, “a 10 percent change in RRC yields a 2 percent change in vehicle fuel economy.” This approach is generally acceptable for the relatively narrow range of RRCs observed for most passenger tires. However, it can introduce imprecision, since a given percentage change in fuel economy is linearly related to an absolute change in rolling resistance. As RRC becomes smaller, a given percentage reduction in RRC tends to have a diminished effect on fuel economy. Nevertheless, because percentage change is a common and widely understood concept, it is often used in this report. Many studies have examined the relationship between rolling resistance and fuel economy. A comprehensive review of fuel economy data from more than a dozen studies published before 1990 was undertaken by Schuring and Futamura (1990). The authors found a narrow range of results that suggested an approximately linear relationship between changes in rolling resistance and fuel economy. During the time period of the studies reviewed, new-tire RRCs were seldom lower than 0.01, so that a 10 percent differential was equivalent to a difference in

RRC of 0.001 or more. For passenger cars operated in urban environments characterized by stop-and-go driving, a 10 percent reduction in the average RRC for all tires on a vehicle was found to increase fuel economy by 1.2 to 1.5 percent. For highway driving characterized by higher and more consistent travel speeds, the same percentage reduction in RRC increased fuel economy by 0.9 to 2.1 percent. Estimates of the fuel economy response for combined urban and highway driving schedules varied from 1.15 to 2.1 percent per 10 percent change in RRC. While fewer studies were performed on light-duty trucks, their corresponding fuel economy effects ranged from 0.95 to 1.25 percent for combined urban and highway driving.

The findings of Schuring and Futamura were consistent with common assumptions and rules of thumb concerning the fuel economy response to changes in rolling resistance. For instance, Thompson and Reineman (1981), in assisting EPA with the development of fuel economy models, assumed that a change of 0.001 in RRC would change vehicle fuel consumption by 1 percent during urban driving and 2 percent during highway driving. Studies published more recently have yielded similar results. Schuring (1994) estimated that for passenger tires having an RRC of 0.012, a 10 percent reduction in RRC will cause fuel economy to increase by 1.4 percent on average—and within a range of 0.7 to 2 percent, depending on the tire’s duty cycle and operating conditions. Schuring found the relationship to be approximately linear. He calculated that the theoretical limit for fuel savings—that is, under the hypothesis that rolling resistance could be eliminated entirely—is 14 percent for conventional passenger cars and 28 percent for fully loaded large trucks. More recently, Hall and Moreland (2001, 527) assumed a more conservative 0.5 to 1.5 percent increase in fuel economy per 10 percent reduction in RRC, although they did not give the baseline RRC.

In interviews with original equipment manufacturers (OEMs) for this study (as discussed in more detail later in the chapter), one—General Motors—permitted the use of its CAFE simulation model to predict fuel consumption effects from changes in RRC. The committee commissioned Environmental Energy Analysis, Inc. (EEA), to run and review the simulations, including the fuel economy model of the National Energy Technology Laboratory (NETL). In addition, Professor Marc Ross of the

University of Michigan provided the study committee with estimates of the fuel consumption effects derived from a computational model. All of the models are based on a four-cylinder, gasoline-powered midsize passenger car. The results of the simulations are given in Table 3-1. They too assume a 10 percent change in RRC, but from a conservatively smaller base coefficient of 0.008—meaning an incremental change in RRC of ±0.0008.

The results of literature reviews and the output of these simulations are sufficiently consistent to estimate a response range for RRC that is meaningful for most driving patterns and common types of passenger vehicles. They are consistent with the long-standing rule of thumb that a 10 percent reduction in RRC will yield a 1 to 2 percent increase in vehicle fuel economy. The lower end of the 1 to 2 percent range, however, is more relevant for tires having low RRCs and driven in urban environments, while the higher end is more relevant for tires having higher RRCs and driven on highways. As explained above, each percentage reduction in RRC becomes smaller in absolute terms. Hence, a more precise way to state the fuel economy response is that each 0.001 reduction in RRC causes fuel economy to increase by 1 to 2 percent.

In support of its growing array of regulatory programs concerning motor vehicle emissions and fuel economy, the federal government began paying

attention to tire rolling resistance in the 1970s.⁷ When J1269 was issued in 1979, EPA was one of the first organizations to use it to test new passenger tires. Having observed a large positive effect on fuel economy from the mass introduction of radial-ply tires, EPA suspected that variations in the rolling resistance of tires installed on new vehicles could have measurable effects on both emissions and fuel economy test results. The agency therefore began testing common passenger tires for rolling resistance to ascertain the magnitude of this effect.

During the 25 years since EPA tested a 54-tire sample of bias- and radialply passenger tires, few additional data on tire rolling resistance have become publicly available for either replacement or OE passenger tires. The publicly available data sets are reviewed below, beginning with EPA’s 54-tire sample from 1982 and 1983 and continuing with more recent information from Consumer Reports, private research consultants, and submissions to NHTSA and U.S. Department of Transportation rule-making. Of most significance, rolling resistance measurements for more than 150 new passenger tires were made publicly available by three major tire companies during the course of this study. While the data set has limitations, it contains data on many new tires currently on the market and supplemental data on each tire’s speed rating, size, traction and tread wear ratings, tread depth, and retail price.

Unless otherwise specified, all RRCs in the data sets discussed below were derived by using the J1269 procedure on new tires. For reasons given in the Appendix, the committee is confident that this test procedure leads to ordinal rankings of tires in terms of rolling resistance that are comparable with those that would be expected from applying the J2452 procedure. Because it has been used for more than 25 years, the J1269 procedure allows for comparisons of RRC measurements across data sets that span two decades or more. However, as with all testing conducted at different times, by different laboratories, and with different equipment, some of the observed variability in RRCs—both across and within data sets—may be attributable to the testing mechanisms themselves. The committee acknowledges this potential but has no reason to believe that any testing

discrepancies would follow a particular pattern or be of a magnitude that would severely compromise general comparisons across data sets.

The RRC measurements from several of the data sets discussed in this section are presented in tables. The specific values are shown because some of the data sets are unpublished or their original sources are difficult to obtain. The RRC data supplied by the Rubber Manufacturers Association (RMA), however, are too lengthy to provide in this report. The RMA data accompany the downloadable version of this report at the TRB website location trb.org/news/blurb_detail.asp?id=5973.

EPA conducted the first government-sponsored measurements of newtire rolling resistance during the early 1980s with the SAE J1269 test procedure (Thompson and Reineman 1981; Egeler 1984). The agency funded testing of 252 individual tires from 20 manufacturer brand names and 54 model lines. The sample consisted of 36 radial-ply and 18 bias-ply lines. The tires were sampled to be representative of the most popular tires at that time and were believed to include more than half of the tire lines in the replacement market. All of the tires tested were P195/75 with 14-inch rim diameters, which was believed to be the most common size at the time. Four to six tires were tested from each tire line to calculate an average rolling resistance for each of the 54 models.

Table 3-2 presents the results only for the 36 radial-ply lines tested by EPA. The results for the 18 bias-ply tires, which are no longer in common use, are omitted. All of the tires were tested when new, after the break-in protocols of the SAE test procedure were followed. The reported RRCs for the 36 tires ranged from 0.0098 to 0.0138, with a mean of 0.0113. In every case but one, the radial-ply tires exhibited significantly lower new-tire rolling resistance than the bias-ply lines. The average RRC for radial-ply tires was more than 20 percent lower than that of the bias-ply group.

In 1994, when NHTSA proposed adding a fuel economy label for passenger tires as part of the Uniform Tire Quality Grading (UTQG)

system,⁸ most tire companies opposed the proposal in comments submitted to the agency (NHTSA 1995). Michelin was the only major tire company to approve of the proposed addition to the UTQG. In its initial comments to NHTSA, Michelin reported RRCs for nine OE and 37 replacement tires measured when they were new.⁹ The 46 tires were from a variety of lines manufactured by Michelin, Bridgestone, Cooper, Goodyear, and other tire makers (Table 3-3). The basis for the sample was not given, nor were the tire sizes. The nine OE tires had an average RRC of 0.0091 and fell within a range from 0.0073 to 0.0105. The 37 replacement tires had an average RRC of 0.0112 and fell within a range from 0.0087 to 0.0143.

In a submission to NHTSA the following year (1995),¹⁰ Michelin provided the rolling resistance specifications for 24 OE tires that were supplied to 10 automobile manufacturers for several Model Year 1995 vehicles. Again, the RRCs were measured when the tires were new. The values ranged from 0.0077 to 0.0114 (Table 3-3). Michelin also tested replacement tires from six tires lines, including three consisting of P215/70/R15 tires and three consisting of P235/75/R15 tires. The tires were from Michelin, Goodyear, and Continental. The RRCs for the six tire lines ranged from 0.0089 to 0.0128.

In other comments to NHTSA in the same rulemaking, Goodyear provided its own estimates of the range of RRCs commonly found among OE and replacement tires. It estimated ranges of 0.0067 to 0.0152 for new OE tires and 0.0073 to 0.0131 for new replacement tires, although it did not name the tires included.¹¹

Since its initial rolling resistance tests in the early 1980s, EPA has performed additional work on tire energy performance, mainly in support

of its climate change programs. In 2001, it conducted load and fuel economy tests on several tires installed on the same vehicle (Automotive Testing Laboratories 2002). The results of this work are presented here for informational purposes only. The agency intended to use the test results to develop a tire ranking system for rolling resistance, to be made available on its website or in a “Green Car Guide.”¹² Because EPA did not measure RRCs for the tires tested, the data are difficult to compare with other measurement data and are not referred to again in this report.

Five Model Year 2001 passenger vehicles (Dodge Caravan, Ford F150, Chevrolet Suburban, Toyota Camry, and Honda Civic) were tested when equipped with their original tires and with popular replacement tires. None of the OE tires was new; each set had been in service between 2,000 and 14,700 miles. Four of the vehicles were tested with one set of new replacement tires, and a fifth vehicle (Camry) was tested with five sets of new replacement tires.

Two separate tests were carried out to measure the forces associated with tires. The first test, designed to measure variations in road load, was conducted on a 7.5-mile test track. Each vehicle was driven to a speed of 125 km/h and then placed in neutral to coast. Deceleration was recorded at various intervals to calculate road load forces. Higher deduced loads were assumed to be indicative of higher rolling resistance. At the 50-mph

interval, the road load force was 4.4 to 14.3 percent higher for the replacement tires on four of the tested vehicles. In the case of the Camry, four of the five replacement tires exhibited lower road load forces than did the OE tires (and presumably lower rolling resistance), by 0.2 to 10.6 percent. No explanation of why the Camry results differed from those of the other vehicles was offered.

EPA conducted coastdown measurements for each tire group on a chassis dynamometer. The resistance forces at 50 mph showed a similar pattern; the replacement tires measured higher loads by 2 to 5.7 percent. The exception was the Camry. For that vehicle, the measured rolling resistance of four of the five replacement tires was lower than that of the OE tires by 13 to 26 percent.

EPA also tested the vehicles and their tire groupings for fuel economy by using the federal test procedure. Measurements of fuel economy were lower by 0.5 to 5.5 percent in four of the five vehicles when equipped with replacement tires. Meanwhile, measurements of vehicle fuel economy for the Camry were higher by 1.3 to 10.4 percent for four of the five replacement tires.

With funding from the Energy Foundation, Ecos Consulting—a private consulting organization—sponsored tests measuring the rolling resistance of 48 new replacement tires during 2002. The tires were selected to cover the products of several manufacturers and to include a mix of sizes and types. The rolling resistance measurements were conducted under the SAE J1269 test procedure. The 48 tires originally included seven light truck (LT-metric) and seven specialty winter tires. These 14 tires are excluded from the data set as it is examined here, given this study’s focus on passenger tires. The 34 remaining passenger tires consisted of four groupings of sizes: P185/70R14, P235/75R15, P205/55R16, and P245/75R16. About one-third were from performance lines (H-rated and above). The RRC measurements are shown in Table 3-4. They range from 0.00615 to 0.01328, with an average of 0.0102 and a median of 0.0104.

Because this data set is contemporary and the tire names and sizes are identified, the committee was able to collect supplemental information for each tire, including its UTQG system grades, tread depth, and retail

prices. The data are analyzed later in the report, along with tire data from other sources.

Consumers Union periodically tests categories of passenger tires for various performance attributes of interest to consumers and publishes the results in Consumer Reports. In recent years, it has tested passenger tires commonly used on SUVs and pickup trucks (November 2004) and performance tires used mainly on passenger cars (speed rated H and above) (November 2003). A total of 40 tires were tested, including 22 all-season SUV/pickup tires and 18 performance-rated tires. The 22 SUV/pickup tires were all size P235/70/R16 with speed ratings of S or T. The sizes of the 18 performance tires were not given.

Presumably, rolling resistance was measured when the tires were new, although Consumer Reports did not report the rolling resistance values derived from the tests or the exact test procedures used—except to note that measurements were taken on a dynamometer at 65 mph. The results were presented in a qualitative manner in Consumer Reports. Of the 40 tires tested, the rolling resistance of 21 was characterized as excellent or very good, 15 as good or fair, and 4 as poor. Consumer Reports stated that the difference in vehicle fuel economy (miles per gallon) between a tire rated as excellent and one rated as poor is about 2 percent at 65 mph. Results from multiple sizes within a tire line were not given.

The Consumer Reports results are not examined further in this study because of the qualitative nature of the ratings information. Some of the tires tested by Consumers Union (including the exact sizes) also appear in the RMA data set discussed later. From this limited comparison of the two data sets, it appears that Consumer Reports characterizes tires with RRCs below 0.01 as excellent, between 0.01 and 0.011 as very good or good, and above 0.011 as poor.

Interested in learning more about the rolling resistance characteristics of OE tires, committee members and staff interviewed representatives from several OEMs: General Motors, Daimler Chrysler, and Ford Motor

Company. The interviews yielded information on rolling resistance values and ranges for new OE tires, as well as projected effects of incremental changes in tire rolling resistance on motor vehicle fuel economy. The meetings also provided insights into OEM expectations about future trends in rolling resistance and the relationship between rolling resistance and other tire performance characteristics, which are discussed later in this report. Because the discussions with the OEMs involved proprietary information, the committee agreed not to disclose the identity of individual companies giving specific information.

As has been noted, all automobile manufacturers maintain staff with tire expertise and have tire testing capabilities. Rolling resistance is an important consideration in specifying tires for most vehicle models, but specifications differ by vehicle and by tire depending on the other performance capabilities of interest for the vehicle class and type. As a preface to their comments, all three OEMs emphasized that the resulting balance of performance attributes changes over time as tire technologies improve. All have observed progressive improvements in tire properties over time; consequently, comparisons of tires at different technology levels may not reveal the same pattern of trade-offs required to achieve a specific balance of capabilities and tire supply costs.

When asked to approximate the range of rolling resistance values specified for their new tires, the OEMs noted that their individual ranges may differ in part because of variability in tire testing equipment, applied correction factors, and the reference conditions used in calculating and reporting specific RRCs. They cautioned that this variability alone could result in RRC differentials of as much as ±20 percent among the ranges reported by each company and in comparison with RRCs observed among replacement tires. All of the OEMs reported measuring and specifying rolling resistance under the SAE J2452 test procedure because the results can be fitted into models for the federal driving cycles used in emissions and fuel economy testing. Achieving federal emissions and fuel economy targets is a major reason why OEMs are concerned with rolling resistance.

One of the OEMs indicated that the following new-tire rolling resistance values are typical for four general categories of OE passenger tires:

• All-season, 0.007;

• Touring, 0.008;

• Performance, 0.01; and

• Light truck passenger, 0.0075 to 0.0095.

The all-season and touring tires are the most common tires installed on its passenger cars, with the latter more common for more expensive and higher trim level cars.

Another OEM provided the following new-tire rolling resistance ranges for similar tire categories, which were derived by using the SAE J2452 test procedure and reported by using the Standard Mean Equivalent Rolling Force conditions described in the Appendix:

• All-season, 0.005 to 0.0062;

• Touring, 0.0058 to 0.0075;

• Performance, 0.0065 to 0.0083; and

• High performance, 0.009.

The company did not provide typical rolling resistance values for light truck passenger tires.

The final OEM did not provide rolling resistance ranges but offered relevant observations with regard to its experience in testing and specifying rolling resistance. It has observed significant changes in the rolling resistance characteristics of a tire during break-in and initial operation. The company has found that tread rubber changes permanently during the first 4,000 miles of use, resulting in lower rolling resistance. Thus, in general, the company relies on vehicle coastdown testing for rolling resistance in evaluating tires for application on its vehicles (similar to the test methods used by EPA in 2001 described above). The company has found this test method to be more reliable for selecting tires that can help achieve vehicle emissions and fuel economy targets, since changes in rolling resistance occur during tire break-in.

In commenting on future tire developments, the OEMs observed that current tire trends are already having mixed effects on rolling resistance. The trends toward larger rim diameters and lower aspect ratios among performance tires are generally helpful in reducing rolling resistance, but they are normally accompanied by the addition of hysteretic material to improve cornering and stopping capabilities, which the OEMs believe may be increasing rolling resistance. Run-flat tires, which are becoming more popular, appear to have at least 20 percent higher rolling resistance

than the conventional OE tires supplied on the same vehicle, in part because run-flat tires have additional structural material and mass. However, one model of run-flat tire was reported to have lower rolling resistance because of its internal bracing, which reduces deformation.

One OEM reported that tires installed on hybrid vehicles are generally not specified any differently from those installed on nonhybrid cars designed to achieve high fuel economy. Another noted that the attention given to tire rolling resistance can be expected to increase with the advent of hybrid drivetrains and technologies such as cylinder cutout, since the fuel economy effects are greater. In some cases, low-rolling-resistance tires have enabled increases in the operating range of cylinder cutout.

Through RMA, three major tire manufacturers—Michelin, Goodyear, and Bridgestone—provided the committee with rolling resistance measurements, UTQG system grades, and speed ratings for 162 passenger tires of varying sizes and affiliated brands (e.g., Uniroyal, Firestone, BFGoodrich). The Michelin portion of the data set consisted of 135 tires from more than three dozen lines in the replacement market. Bridgestone provided data for 24 tires from five lines, including five OE tires. Goodyear data covered 13 tires from four lines, including three OE tires. Originally included among the Michelin data were 44 light truck and winter tires, which the committee excluded from the main data set.

In all cases, the RRC measurements were obtained with the SAE J1269 test procedure. All of the RRCs were derived from measurements of tires tested when new. In providing the data, the three tire companies emphasized that the RRC values reported by each company could exhibit variability in part because of the differences in testing equipment used for RRC measurement. Such testing variability, coupled with the variability in the number and selection of tires reported by each company, precludes comparisons of patterns across tire companies. Hence, the data reported by the three tire companies are examined in the aggregate and are referred to in the following as the RMA data set.

The RMA data set includes tires of many sizes and speed ratings. Table 3-5 contains summary statistics for the data set derived from three

tire manufacturers. Of the 162 tires sampled, 97 (60 percent) are speed rated S or T, 31 (19 percent) are rated for performance (speed rated H or V), and 34 (21 percent) are rated for high performance (speed rated W, Y, or Z). A large majority of the tires (74 percent) have rim diameters of 15, 16, or 17 inches. In addition, three-quarters of the sampled tires have aspect ratios of 60 to 75, while the remaining tires have lower ratios (mostly 45, 50, and 55). Tire section widths range from 175 to 335 millimeters; tires with section widths between 195 and 245 millimeters account for 70 percent of the tires sampled. Among the 162 tires, there are more than 70 distinct size (section width, aspect ratio, and rim diameter) and speed rating (S, T; H, V; W, Y, Z) combinations.

It is difficult to ascertain how representative the 162 tires are of the general population of passenger tires sold each year in the United States. Data on industry shipments suggest that the above data set contains a higher-than-average percentage of performance tires. The RMA Fact-book for 2005 indicates that tires with speed rating S or T accounted for 73 percent of replacement tire shipments in 2004, while tires with higher speed ratings—H or V and W, Y, or Z—accounted for 22 and 4 percent, respectively (RMA 2005, 22).

In addition, the RMA data were provided without information on the sampling methodology. Some of the data points represent single tests on individual tires, and other data represent more than one test. While these shortcomings limit the degree to which definitive findings can be attributed to analyses of the data, the RMA data set is by far the largest single source of publicly available data on rolling resistance for new tires sold in the United States. In this respect, it offers many opportunities for analyzing rolling resistance levels and relationships with respect to other attributes such as wear resistance, traction, size, selling price, and speed ratings. To expand these analytic opportunities, the committee supplemented the information provided by the tire companies with publicly available data on each tire’s tread depth, weight, and retail prices obtained from manufacturer and tire retailer websites. These data are analyzed in Chapters 4 and 5 to assess possible relationships with rolling resistance.

The focus of the remainder of this chapter is on the new-tire rolling resistance values observed in the RMA data. Because the data set contains

FIGURE 3-2 Distribution of tires in the RMA data set by RRC.

only eight tires identified as current OE tires, which is too few for useful comparisons, the emphasis of the statistical assessment is on the 154 replacement tires in the data set.¹³

The range of RRCs observed for the 154 replacement tires in the RMA data set is 0.0065 to 0.0133, with a mean and median of 0.0102 and 0.0099, respectively (Figure 3-2). More than half (55 percent) of the tires have an RRC between 0.009 and 0.011. Coefficients below 0.008 or above 0.013 can be characterized as unusually low or high, and such values occur in less than 8 percent of the tires sampled.

A simple sorting of the data by speed rating reveals that the performance-rated tires have a slightly higher-than-average rolling resistance. The average for S and T tires is 0.0098, while the averages for H, V and W, Y, Z tires are 0.0101 and 0.0113, respectively. This pattern suggests a relationship between RRC and speed rating. However, performance tires are more likely to have lower aspect ratios, wider section widths, and larger rim diameters than tires with lower speed ratings. Thus, geometric differences in tires may contribute to rolling resistance differentials just as much as the design elements intended to augment performance.

A sorting of the data by rim diameter suggests that tire dimensions can indeed have an effect on rolling resistance measurements. Tires with a rim diameter of 15 inches or lower have an average rolling resistance of 0.0106, more than 10 percent above the average of 0.0093 for the tires with a higher rim diameter.

Multivariate statistical analyses are required to control for the many tire design variables that may be related to rolling resistance. Such an analysis is performed in the next chapter to shed light on the full array of relationships between rolling resistance and other tire characteristics such as tread depth and tread wear. Nevertheless, a simple descriptive sorting of the data by tire speed ratings and size dimensions offers some insights into the variations in RRC that occur within groupings of tires having the same size and speed ratings. Figure 3-3 shows the distribution of RRCs for the seven most popular speed rating–size configurations in the RMA data set, which includes 51 of the 154 replacement tires in the data set. The sorting reveals wide ranges in RRCs within such groupings of like tires. In all seven groupings, the difference between the highest and lowest value is at least 18 percent, and most of the differentials exceed 25 percent.

Table 3-6 summarizes the RRCs from the above-referenced data sets, starting with the early EPA data and ending with the RMA data from 2005. As noted, the 1982–1983 EPA measurements confirmed the large reductions in rolling resistance caused by the introduction of radial-ply tires, although

FIGURE 3-3 Distribution of RRCs for tires in the most common size and speed rating groupings, RMA data set.

most RRCs for radial tires in 1982–1983 exceeded 0.01. The Michelin-reported data for replacement tires on the market in the mid-1990s show further progress in reducing rolling resistance, especially in the number of tires achieving RRCs below 0.01. The most recent data, from Ecos Consulting in 2002 and RMA in 2005, reveal additional reductions in the average and median rolling resistance. Nearly 20 percent of the tires sampled in these more recent (2002 and 2005) data sets had rolling resistance measurements of 0.009 or less. In comparison, none of the tires sampled by EPA in the early 1980s, and only two tires in the Michelin-reported data from 1994 and 1995, had an RRC lower than 0.009.

Most notable are the gains made among the top-performing tires with respect to rolling resistance. The 25 percent (or quartile) of tires having the lowest RRCs in the 1982–1983 data set had an average RRC of 0.0103. This compares with an average RRC of 0.0085 for the same quartile for the combined 2002 and 2005 data. Figure 3-4 shows a plot of the RRCs from the various data sets. It displays the persistence of tires at the high

FIGURE 3-4 Rolling resistance values for passenger tire samples, 1982 to 2005.

end of the RRC spectrum in all data sets, across all periods. In 1982– 1983, the quartile of tires with the highest RRCs had an average coefficient of 0.0126. In the combined data for 2002 and 2005, this quartile had comparable RRCs, averaging 0.0125.

A possible explanation for the widening spread in RRCs among today’s tires is the proliferation of tire sizes and speed ratings. The 1982–1983 EPA data are for a single tire size (P195/75/15). In that period, speed ratings were uncommon in North America. Today’s replacement tires—as represented in the 2002 and 2005 data sets—include many high-performance tires. These tires, with speed ratings of W, Y, and Z, account for a disproportionate share of tires with high RRCs, as shown in Figure 3-5. Indeed, they account for most tires having RRCs greater than 0.012, whereas S and T tires (which are not considered performance tires) account for all of the values observed below 0.008. Nevertheless, Figure 3-5 also shows a persistent spread in RRCs, even when rim diameter and speed ratings are controlled for. Speed rating is not the only factor affecting rolling resistance. About one-third of the high-performance tires have RRCs below 0.01, and about 20 percent of the S and T tires have RRCs greater than 0.011.

FIGURE 3-5 Distribution of rolling resistance coefficients in 2002 and 2005 data sets compared with distribution in 1982–1983 EPA data set, controlling for rim size and speed rating.

There is an evident relationship between rim diameter and rolling resistance that warrants closer examination when the combined 2002 and 2005 data are compared with the 1982–1983 EPA data. Many of the S and T tires that have higher RRCs in the 2002 and 2005 data possess rim diameters of 13 and 14 inches. EPA only tested tires with 15-inch rim diameters. Among contemporary tires with 15-inch rim sizes, there are noticeably more with low RRCs than in the EPA data from two decades earlier. The entire distribution appears to have shifted downward by about 10 percent (Figure 3-5). Most of the higher RRCs continue to be found among the tires with smaller 13- and 14-inch rim sizes, nearly all of which are S and T tires.

The average retail price for the 13- and 14-inch S and T tires is about 50 percent ($60) below the average ($117) for all of the tires represented in the data for 2002 and 2005.¹⁴ Hence, it is reasonable to ask whether the RRC distributions observed in this chapter are related in part to unex-

amined factors such as tire construction cost and life expectancy, which may have a strong correlation with other examined variables such as tire size and speed rating. More consideration is given in the following chapters to these and other aspects of tire performance that may have a bearing on rolling resistance.

Most of the energy contained in a tank of motor fuel is dissipated as unrecoverable heat from engine combustion and friction in the driveline. Some of the energy output from the engine powers vehicle accessories. Only about 12 to 20 percent of the energy originating in the fuel tank is ultimately transmitted through the vehicle’s driveline as mechanical energy to turn the wheels. Rolling resistance consumes about one-third of this energy output. Aerodynamic drag and braking consume the remainder. Rolling resistance, therefore, directly consumes a small portion (one-third of the 12 to 20 percent) of the total energy expended by the vehicle.

However, reducing rolling resistance, and thus mechanical energy demand, by a given amount translates into a larger reduction in total fuel consumption because less fuel needs to be sent to the engine. The effect on total fuel consumption will depend on a number of factors, including the efficiency of the engine and driveline as well as the amount of energy used to power accessories. For most passenger vehicles, a 10 percent reduction in average rolling resistance over a period of time will lead to a 1 to 2 percent reduction in fuel consumption during that time.

The main source of rolling resistance is hysteresis, which is caused by the viscoelastic response of the rubber compounds in the tire as it rotates under load. The repeated tire deformation and recovery causes mechanical energy to be converted to heat; hence additional mechanical energy must be supplied to drive the axle. The design characteristics of a tire that affect this energy loss are its construction; geometric dimensions; and materials types, formulations, and volume. The tread, in particular, has a major role in hysteresis because it contains large amounts of viscoelastic rubber material. As tread wears, a tire’s rolling resistance declines, primarily because of the reduction in the amount of viscoelastic material.

Travel speed within the range of normal city and highway driving has relatively little effect on rolling resistance. The main operating conditions

that affect tire hysteresis are load, inflation pressure, alignment, and temperature. The more a tire is loaded at a given pressure, the more it deforms and suffers hysteretic losses. A tire deforms more when it is underinflated. For tires inflated to pressures of 24 to 36 psi, each 1-psi drop in inflation pressure increases the tire’s rolling resistance by about 1.4 percent. This effect is greater for inflation pressures below 24 psi. Consequently, maintenance of tire pressure is important for a tire’s energy performance as well as for tire wear and operating performance.

Rolling resistance is proportional to wheel load and can therefore be measured and expressed in terms of a constant RRC. Thus, tires with low RRCs have low rolling resistance. Standard test procedures have been developed to measure RRC. The vast majority of replacement passenger tires have RRCs within the range of 0.007 to 0.014 when measured new, while the range for new OE tires tends to be lower—on the order of 0.006 to 0.01. Federal fuel economy standards have prompted automobile manufacturers to demand OE tires with lower rolling resistance. Information on precisely how these lower-rolling-resistance characteristics have been achieved is proprietary.

In general, each incremental change in RRC of 0.001 will change vehicle fuel consumption by 1 to 2 percent. Thus, for an average passenger tire having a coefficient of 0.01, a 10 percent change in RRC will change vehicle fuel consumption by 1 to 2 percent. The lower end of the range is more relevant for tires having lower RRCs and operated at lower average speeds, while the higher end of the range is more relevant for tires having higher RRCs and operated at highway speeds.

Today’s passenger tires offer better performance and capability than did previous generations of tires because of continued innovations and refinements in tire design, materials, and manufacturing. Significant progress has been made in reducing rolling resistance—as measured in new passenger tires—over the past 25 years. More tire models today, when measured new, have RRCs below 0.009, and the most energy-efficient tires have coefficients that are 20 to 30 percent lower than the most energy-efficient radial models of 25 years ago. Tires at the higher end of the RRC range, however, have not exhibited the same improvement, which has resulted in a widening spread in RRCs over time. The expansion of the number of tire sizes and speed categories, as well as new tire designs to

meet changing vehicle and service applications (e.g., deep-grooved tread for light truck functional requirements and appearance), has likely contributed to the spread in RRCs. However, even among tires of similar size and speed rating, the difference between the tires with the highest and lowest RRCs often exceeds 20 percent.

Tires with high speed ratings (W, Y, and Z) and tires with smaller (13- and 14-inch) rim diameters account for a large share of tires with high rolling resistance. Whether such patterns are related to differences in other tire characteristics, such as size, traction, and wear resistance, is examined in the next chapter.

NHTSA National Highway Traffic Safety Administration

NRC National Research Council

RMA Rubber Manufacturers Association

TRB Transportation Research Board

Automotive Testing Laboratories. 2002. Coastdown and Fuel Economy for Specific Vehicles and Tires. Contract 68-C-00-126. U.S. Environmental Protection Agency, Dec.

Chang, L. Y., and J. S. Shackelton. 1983. An Overview of Rolling Resistance. Elastometrics, March, pp. 18–26.

Clark, S. K. 1983. A Brief History of Tire Rolling Resistance. In Tire Rolling Resistance, Rubber Division Symposia, Vol. 1 (D. J. Schuring, ed.), American Chemical Society, Akron, Ohio.

Clark, S. K., and R. N. Dodge. 1978. A Handbook for the Rolling Resistance of Pneumatic Tires. Report DOT TSC-78-1. U.S. Department of Transportation, June.

DeRaad, L. W. 1978. The Influence of Road Surface Texture on Tire Rolling Resistance. SAE Paper 780257. Presented at SAE Congress and Exposition, Detroit, Mich., Feb. 27–March 3.

Egeler, N. 1984. Characterization of the Rolling Resistance of Aftermarket Passenger Car Tires. Report EPA-AA-SDSB-84-5. Office of Mobile Sources, Standards Development and Support Branch, U.S. Environmental Protection Agency, Ann Arbor, Mich.

Hall, D. E., and J. C. Moreland. 2001. Fundamentals of Rolling Resistance. Rubber Chemistry and Technology, Vol. 74, pp. 525–539.

LaClair, T. J. 2005. Rolling Resistance. In The Pneumatic Tire (J. D. Walter and A. N. Gent, eds.), National Highway Traffic Safety Administration, Washington, D.C., pp. 475–532.

Mars, W. V., and J. R. Luchini. 1999. An Analytical Model for the Transient Rolling Resistance Behavior of Tires. Tire Science and Technology, Vol. 27, No. 3, July–Sept., pp. 161–175.

Martini, M. E. 1983. Passenger Tire Rolling Loss: A Tread Compounding Approach and Its Tradeoffs. In Tire Rolling Resistance (D. J. Schuring, ed.), American Chemical Society, Akron, Ohio, pp. 181–197.

NHTSA. 1995. Light Vehicle Uniform Tire Quality Grading Standards: Notice of Proposed Rulemaking Preliminary Regulatory Evaluation. Office of Regulatory Analysis Plans and Policy, May.

NRC. 1992. Automotive Fuel Economy: How Far Should We Go? National Academy Press, Washington, D.C.

NRC. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. National Academy Press, Washington, D.C.

Pillai, P. S., and G. S. Fielding-Russell. 1991. Effect of Aspect Ratio on Tire Rolling Resistance. Rubber Chemistry and Technology, Vol. 64, No. 3, pp. 641–647.

RMA. 2005. Factbook 2005: U.S. Tire Shipment Activity Report for Statistical Year 2004. Washington, D.C.

Ross, M. 1997. Fuel Efficiency and the Physics of Automobiles. Contemporary Physics, Vol. 38, No. 6, pp. 381–394.

Schuring, D. J. 1980. The Rolling Loss of Pneumatic Tires. Rubber Chemistry and Technology, Vol. 53, No. 3, pp. 600–727.

Schuring, D. J. 1994. Effects of Tire Rolling Loss on Vehicle Fuel Consumption. Tire Science and Technology, Vol. 22, No. 3, pp. 149–161.

Schuring, D. J., and S. Futamura. 1990. Rolling Loss of Pneumatic Highway Tires in the Eighties. Rubber Chemistry and Technology, Vol. 62, No. 3, pp. 315–367.

Small, K., and K. Van Dender. 2005. Fuel Efficiency and Motor Vehicle Travel: The Declining Rebound Effect. Economic Working Paper 05-06-03 (revised December).

Sovran, G., and D. Blaser. 2003. A Contribution to Understanding Automotive Fuel Economy and Its Limits. SAE Paper 2003-01-2070.

Thompson, G. D., and M. E. Reineman. 1981. Tire Rolling Resistance and Fuel Consumption. SAE Paper 810168. Presented at International Congress and Exposition, Detroit, Mich., Feb. 23–27.

TRB. 1997. Special Report 251: Toward a Sustainable Future: Addressing the Long-Term Effects of Motor Vehicle Transportation on Climate and Ecology. National Research Council, Washington, D.C.

Velinsky, S. A., and R. A. White. 1979. Increased Vehicle Energy Dissipation due to Changes in Road Roughness with Emphasis on Rolling Loss. SAE Paper 790653.