Tyre Paper PDF

DOT HS 811 154 August 2009
NHTSA Tire Fuel Efficiency

Consumer Information
Program Development:
Phase 2 – Effects of Tire Rolling
Resistance Levels on Traction,
Treadwear, and Vehicle Fuel Economy
This document is available to the public from the National Technical Information Service, Springfield, Virginia 22161
This publication is distributed by the U.S. Department of Transportation, National Highway
Traffic Safety Administration, in the interest of information exchange. The opinions, findings and
conclusions expressed in this publication are those of the author(s) and not necessarily those
of the Department of Transportation or the National Highway Traffic Safety Administration.
The United States Government assumes no liability for its content or use thereof. If trade or
manufacturers’ names or products are mentioned, it is because they are considered essential
to the object of the publication and should not be construed as an endorsement. The United
States Government does not endorse products or manufacturers.
TECHNICAL REPORT DOCUMENTATION PAGE
1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.

DOT HS 811 154
4. Title and Subtitle 5. Report Date
NHTSA Tire Fuel Efficiency Consumer Information Program Development: Phase 2 – August 2009
Effects of Tire Rolling Resistance Levels on Traction, Treadwear, and Vehicle Fuel 6. Performing Organization Code
Economy
7. Author(s) 8. Performing Organization Report No.
Larry R. Evans,1 James D. MacIsaac Jr.,2 John R. Harris,1 Kenneth Yates,2 Walter
Dudek,1 Jason Holmes,1 Dr. James Popio,3 Doug Rice,3 Dr. M. Kamel Salaani1
1
Transportation Research Center, Inc., 2National Highway Traffic Safety Administra
tion, 3Smithers Scientific Services, Inc.
9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)
National Highway Traffic Safety Administration
Vehicle Research and Test Center 11. Contract or Grant No.
P.O. Box B-37 DTNH22-03-D-08660,

10820 State Route 347 DTNH22-07-D-00060
East Liberty, OH 43319-0337
12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered
National Highway Traffic Safety Administration Final
1200 New Jersey Avenue SE. 14. Sponsoring Agency Code
Washington, DC 20590 NHTSA/NVS-312

15. Supplementary Notes
Project support and testing services provided by: NHTSA San Angelo Test Facility, Akron Rubber Development
Laboratory, Inc., Smithers Scientific Services, Inc., Standards Testing Laboratories, Inc., and Transportation Research
Center, Inc.
16. Abstract
This report summarizes the second phase of the project to develop a tire fuel efficiency consumer information program
intended to examine possible correlations between tire rolling resistance levels and service variables such as vehicle fuel
economy, wet and dry traction, and outdoor and indoor treadwear. Tires of 15 different models with known rolling
resistances were installed on the same new passenger car to evaluate their effects of on vehicle fuel economy. A 10
percent decrease in tire rolling resistance resulted in an approximately 1.1-percent increase in fuel economy for the
vehicle. This result was within the range predicted by technical literature. Reducing the inflation pressure by 25 percent
resulted in a small but statistically significant decrease of approximately 0.3 to 0.5 miles per gallon for four of the five
fuel economy cycles, excluding the high-speed, high-acceleration US06 cycle. This value was smaller than many values
predicted by technical literature, and possible explanations are being explored.
Tires of 16 different models with known rolling resistances were subjected to dry and wet skid-trailer testing on asphalt
and concrete skid pads. Both the peak (maximum) and slide (fully locked-tire) coefficients of friction were measured
and indexed against the control tire. For the tires studied, there appeared to be no significant relationship between dry
peak or slide numbers and rolling resistance. However, these tire models exhibited a strong and significant relationship
between better rolling resistance and poorer wet slide numbers. The peak wet slide number displayed the same
tendency, but the relationship was much weaker. This may be significant to consumers without anti-lock braking
systems (ABS) on their vehicles since the wet slide value relates most closely to locked-wheel emergency stops. For
newer vehicles with ABS or electronic stability control systems, which operate in the earlier and higher wet peak
friction range, the tradeoff is less significant. For the subset of 5 tire models subjected to on-vehicle treadwear testing
(UTQGS), no clear relationship was exhibited between tread wear rate and rolling resistance levels. For the subset of 6
tire models subjected to significant amounts of wear in the indoor treadwear tests, there was a trend toward faster wear
for tires with lower rolling resistance. This report concludes with an analysis of the various options in the draft ISO
28580 rolling resistance test and their likelihood of inducing variability in the test results, as well as a discussion of data
reporting format.
17. Key Words 18. Distribution Statement
Tire, rolling resistance, consumer information, tire traction, Energy This report is free of charge from the NHTSA
Independence and Security Act of 2007 (EISA) Web site at www.nhtsa.dot.gov
19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of Pages 22. Price
Unclassified Unclassified 153
Form DOT F 1700.7 (8-72) Reproduction of completed page authorized
i
Approximate Conversions to Metric Measures Approximate Conversions to English Measures
Symbol When You Know Multiply by To Find Symbol Symbol When You Know Multiply by To Find Symbol
LENGTH LENGTH
in inches 2.54 centimeters cm mm millimeters 0.04 inches in

ft feet 30 centimeters cm cm centimeters 0.4 inches in
mi miles 1.6 kilometers km m meters 3.3 feet ft
km kilometers 0.6 miles mi
AREA
AREA
in2 square inches 6.5 square centimeters cm2
ft2 square feet 0.09 square meters m2 cm2 square centimeters 0.16 square inches in2
mi2 square miles 2.6 square kilometers km2 km2 square kilometers 0.4 square miles mi2
MASS (weight) MASS (weight)
oz ounces 28 grams g g grams 0.035 ounces oz

lb pounds 0.45 kilograms kg kg kilograms 2.2 pounds lb
PRESSURE PRESSURE
psi pounds per inch2 0.07 bar bar bar bar 14.50 pounds per inch2 psi
psi pounds per inch2 6.89 kilopascals kPa kPa kilopascals 0.145 pounds per inch2 psi
VELOCITY VELOCITY
mph miles per hour 1.61 kilometers per hour km/h km/h kilometers per hour 0.62 miles per hour mph
ACCELERATION ACCELERATION
ft/s2 feet per second2 0.30 meters per second2 m/s2 m/s2 meters per second2 3.28 feet per second2 ft/s2
TEMPERATURE (exact) TEMPERATURE (exact)
F Fahrenheit 5/9 (Celsius) - 32C Celsius C C Celsius 9/5 (Celsius) + 32F Fahrenheit F
ii
TABLE OF CONTENTS
1.0 INTRODUCTION ................................................................................................................................ 1
1.1 THE CONCEPT OF ROLLING RESISTANCE .................................................................................... 3
2.0 METHODOLOGY ............................................................................................................................... 7
2.1 TEST TIRES .................................................................................................................................. 7
2.1.1 ASTM F2493 Radial Standard Reference Test Tire ............................................................... 7
2.2 TIRE ROLLING RESISTANCE TEST PROCEDURES ......................................................................... 8
2.2.1 ISO Draft International Standard 28580 Single-Point Rolling Resistance ........................... 11
2.2.2 SAE J1269 & ISO 18164 Multi-Point Rolling Resistance.................................................... 11
2.2.3 SAE J2452 Multi-Point (Speed Coast Down) Rolling Resistance........................................ 11
2.3 FUEL ECONOMY TEST VEHICLE ................................................................................................ 11
2.4 TEST WHEELS ............................................................................................................................ 11
2.5 TEST MATRIX ............................................................................................................................ 12
2.6 TREAD COMPOUND PROPERTIES TESTING................................................................................. 13
2.7 ON-VEHICLE FUEL ECONOMY TESTING .................................................................................... 16
2.7.1 EPA 40 CFR Part 86 Dynamometer Fuel Economy Testing ................................................ 17
2.8 SKID-TRAILER TIRE TRACTION TESTING .................................................................................. 21
2.9 ON-VEHICLE TIRE TREADWEAR TESTING ................................................................................. 23
2.10 INDOOR TIRE TREADWEAR TESTING ......................................................................................... 25
3.0 RESULTS............................................................................................................................................ 28
3.1 EFFECT OF TIRE ROLLING RESISTANCE ON AUTOMOBILE FUEL EFFICIENCY ........................... 28
3.1.1 Preliminary Analysis: Data Shifts ......................................................................................... 30
3.1.2 Highway FET Triplicate Analysis: ....................................................................................... 31
3.1.3 Air Conditioning SC03 – 11/20/08 to 11/25/08 .................................................................... 34
3.1.4 Analysis by Date for Possible Drift in Data over Time ........................................................ 35
3.1.5 Effect of Tire Rolling Resistance on Fuel Economy............................................................. 37
3.1.6 Effect of Reduced Inflation Pressure on Fuel Economy ....................................................... 43
3.1.7 Fuel Economy Testing Summary.......................................................................................... 50
3.2 CORRELATION OF TANGENT Δ AT 60°C TO TIRE ROLLING RESISTANCE .................................. 51
3.3 EFFECT OF TIRE ROLLING RESISTANCE ON SAFETY.................................................................. 53
3.3.1 Dry Traction Data ................................................................................................................. 53
3.3.2 Wet Traction Data ................................................................................................................. 56
3.3.3 UTQGS Traction Grade ........................................................................................................ 59
3.3.4 Correlation of Tangent δ at 0°C to Wet Traction Properties................................................. 61
3.4 EFFECTS OF TIRE ROLLING RESISTANCE ON TREADWEAR RATE .............................................. 62
3.4.1 Analysis of Wear Data From Indoor Treadwear Testing ...................................................... 65
4.0 CONCLUSIONS................................................................................................................................. 78
5.0 REQUIREMENTS ............................................................................................................................. 79
6.0 ROLLING RESISTANCE (Fr) VERSUS ROLLING RESISTANCE COEFICIENT Cr)......... 85
6.1 THEORY OF FR AND CR .............................................................................................................. 85
6.1.1 Using Cr from a Single-Load Test to Predict Rolling Resistance at Any Load.................... 90
6.2 DISCUSSION ............................................................................................................................... 94
iii
Appendix 1. Tire and Rim Association, Inc. - Maximum Load Formula for “P” Type Tires..... 100
Appendix 2. Detailed Test Matrix ..................................................................................................... 101
Appendix 3. Examples of Data Acquired From Indoor Treadwear Test ...................................... 103
Appendix 4. Raw Dry Traction Testing Results - Asphalt.............................................................. 124
Appendix 5. Raw Dry Traction Testing Results - Concrete ........................................................... 126
Appendix 6. Raw Wet Traction Testing Results - Asphalt ............................................................. 128
Appendix 7. Raw Wet Traction Testing Results - Concrete........................................................... 130
Appendix 8. UTQG Adjusted Wet Traction Testing Results ......................................................... 132
Appendix 9. ASTM E501 Reference Tire Wet Traction Testing Results ...................................... 134
Appendix 10. ASTM E501 Reference Tire Dry Traction Testing Results .................................. 135
iv
LIST OF FIGURES
Figure 1. Where Does the Energy Go? ........................................................................................... 4
Figure 2. Contribution of Tire Rolling Resistance to Vehicle Fuel Economy Versus Speed......... 5
Figure 3. Magic Triangle: Traction, Treadwear, and Rolling Resistance....................................... 6
Figure 4. Force Method Rolling Resistance Test Machine........................................................... 10
Figure 5. Torque Method Rolling Resistance Test Machine ........................................................ 10
Figure 6. Sample TGA Weight Loss Curve.................................................................................. 14
Figure 7. Tan  as a Function of Temperature From the Tension Test ........................................ 15
Figure 8. Tan  as a Function of Temperature From the Shear Test ............................................ 16
Figure 9. Vehicle Fuel Economy Dynamometer Testing ............................................................. 18
Figure 10. NHTSA San Angelo Skid-Trailer ............................................................................... 22
Figure 11. UTQGS Treadwear Course ......................................................................................... 25
Figure 12. Indoor Treadwear Equipment...................................................................................... 27
Figure 13. Vehicle Fuel Economy Dynamometer Exhaust Collection Bags and Control System32
Figure 14. Highway FET Schedule Fuel Economy Versus Bag Collection Number................... 32
Figure 15. Air Conditioning SC03 Fuel Economy Versus Tire Rolling Resistance by Analysis
Group .................................................................................................................................... 35
Figure 16. Rolling Resistance of Tires Tested Versus Day of Testing......................................... 36
Figure 17. Highway FET (Bag #1) Mileage Versus Tire Rolling Resistance .............................. 39
Figure 20. City FTP Mileage Versus Tire Rolling Resistance ..................................................... 40
Figure 21. High Speed US06 Mileage Versus Tire Rolling Resistance ....................................... 41
Figure 22. Air Conditioning SC03 Mileage Versus Tire Rolling Resistance............................... 41
Figure 23. Cold City FTP Mileage Versus Tire Rolling Resistance............................................. 42
Figure 24. Percentage Change in Fuel Economy Versus Percentage Change in.......................... 43
Figure 25. Tire to Dynamometer Roller Contact / 2008 Chevrolet Impala LS Engine ................ 46
Figure 26. Highway FET (Bag #1) Fuel Economy by Tire Type and Inflation Pressure............. 47
Figure 29. City FTP Fuel Economy by Tire Type and Inflation Pressure.................................... 48
Figure 30. High Speed US06 Fuel Economy by Tire Type and Inflation Pressure...................... 49
Figure 31. Air Conditioning SC03 Fuel Economy by Tire Type and Inflation Pressure ............. 49
Figure 32. Cold City FTP Fuel Economy by Tire Type and Inflation Pressure ........................... 50
Figure 33. Highway FET (Bag #2) Fuel Economy Versus Tire Rolling Resistance by Tire Type
and Inflation Pressure ........................................................................................................... 51
Figure 34. ISO 28580 Rolling Resistance (lbs)Versus Tangent δ at 60°C by Tire Type ............. 52
Figure 35. Dry Traction Numbers Versus ISO 28580 Rolling Resistance ................................... 55
Figure 36. Dry Traction Ratios to E501 Course Monitoring Tire Versus Rolling Resistance ..... 56
Figure 37. Wet Traction Numbers Versus ISO 28580 Rolling Resistance................................... 58
Figure 38. Wet Traction Ratios to E501 Course Monitoring Tire Versus Rolling Resistance..... 59
Figure 39. UTQG Adjusted Traction Coefficient for Asphalt Versus ISO 28580 Rolling
Resistance ............................................................................................................................. 60
Figure 40. UTQG Adjusted Traction Coefficient for Concrete Versus ISO 28580 Rolling
Resistance ............................................................................................................................. 61
v
Figure 41. Slide Traction Number on Wet Concrete Versus Tangent δ at 0°C Measured in
Tension.................................................................................................................................. 62
Figure 42. Projected Tire Mileage to Wearout (Average and Minimum) Versus ISO 28580
Rolling Resistance ................................................................................................................ 64
Figure 43. Average and Fastest Treadwear Rate Versus ISO 28580 Rolling Resistance............. 65
Figure 44. Projected Tire Lifetime for Indoor Treadwear Test .................................................... 67
Figure 45. Treadwear Rate for Indoor Treadwear Test ................................................................ 68
Figure 46. Projected Tire Lifetime for Indoor Treadwear Test .................................................... 73
Figure 47. ISO 28580 Rolling Resistance Versus Tire Weight Loss ........................................... 74
Figure 48. Rolling Resistance as Percent of the Original Rolling Resistance .............................. 75
Figure 49. Percentage of Original Rolling Resistance.................................................................. 77
Figure 50. Temperature Correction Factor - ISO 28580............................................................... 83
Figure 51. Drum Diameter Correction Factor - ISO 28580.......................................................... 84
Figure 52. SAE J1269 Recommended Test - Evaluates Response of Rolling Resistance Force
Over a Range of Three Pressures and Two Loads................................................................ 87
Figure 53. ISO 18164 Annex B - Response of Rolling Resistance Force (Fr) Over a Range of
Three Speeds, Two Pressures, and Two Loads..................................................................... 88
Figure 54. ISO 28580 Test Conditions for Standard Load Passenger Tires................................. 89
Figure 55. Theoretical Single-Load Rolling Resistance (Fr)........................................................ 90
Figure 56. Theoretical Single-Load Rolling Resistance Coefficient (Cr) .................................... 91
Figure 57. Rolling Resistance of 16 Passenger Tires ................................................................... 92
Figure 58. Rolling Resistance Coefficient of 16 Passenger Tires ................................................ 93
Figure 59: Rolling Resistance Force (SAE J1269 Single-Point, Pounds) .................................... 97
Figure 60: Rolling Resistance Coefficient (SAE J1269) .............................................................. 97
vi
LIST OF TABLES
Table 1. 2005 Motor Vehicle Crash Data From FARS and GES, Crashes by Weather Condition 6
Table 2. Phase 2 Tire Models ......................................................................................................... 8
Table 3. Test Matrix...................................................................................................................... 13
Table 4. Analysis of Tread Composition by TGA........................................................................ 14
Table 5. DMA Results for Tangent  at 0C and 60C ................................................................ 16
Table 6. 2008 EPA Fuel Economy 5-Driving Schedule Test (Source: EPA, 2009)..................... 19
Table 7. Fuel Economy Test Schedules........................................................................................ 21
Table 8. Phase 2 Wet and Dry Skid-Trailer Test Tires................................................................. 23
Table 9. On-Vehicle Treadwear Testing....................................................................................... 24
Table 10. Indoor Treadwear Testing............................................................................................. 26
Table 11. Test Parameters............................................................................................................. 26
Table 12. Test Matrix by Date ...................................................................................................... 29
Table 13. Events Identified as Possible Data Shift Correlates...................................................... 31
Table 14. Analysis of Variance for Highway FET Fuel Economy by Tire Type and Collection
Bag Number .......................................................................................................................... 33
Table 15. Air Conditioning SC03 Schedule, mpg for SRTT Tire by Date................................... 34
Table 16. Change in Fuel Economy Over Total Time of Testing ................................................ 36
Table 17. Data Excluded from Fuel Economy Analyses.............................................................. 37
Table 18. ANOVA Results for Effect of Tire Rolling Resistance on Fuel Economy .................. 38
Table 19. Percentage Change in Fuel Economy Versus Percentage Change in Tire Rolling
Resistance ............................................................................................................................. 38
Table 20. Predicted Change in Fuel Economy for 1 psi Change in Tire Inflation Pressure......... 44
Table 21. ANOVA Results for Effect of Tire Inflation Pressure Reduction on Fuel Economy... 50
Table 22. Correlation of Rolling Resistance to Tangent δ at 60°C .............................................. 52
Table 23. Correlation of Properties to Rolling Resistance ........................................................... 53
Table 24. Dry Traction Results, Traction Number and Ratio to E501 Reference Tire ................ 54
Table 25. Pearson Product Moment Correlation of Dry Traction to Rolling Resistance ............. 54
Table 26. Wet Traction Results, Traction Number and Ratio to E501 Reference Tire................ 57
Table 27. Pearson Product Moment Correlation of Wet Traction to Rolling Resistance............. 57
Table 28. Pearson R Product Moment Correlation of Wet Traction to ........................................ 62
Table 29. Analysis of Tire Wear Data .......................................................................................... 63
Table 30. Wear Rates and Projected Mileage to 2/32nds Tread Depth .......................................... 64
Table 31. Indoor Treadwear Tire Wear Data................................................................................ 66
Table 32. Projected Mileage to 2/32nds Inch of Tread Depth ........................................................ 66
Table 33. Projected Lifetime Versus Rolling Resistance – Mild Wear at Tread Center .............. 69
Table 34. Projected Lifetime Versus Rolling Resistance – Severe Wear at Tread Center........... 70
Table 35. Projected Lifetime Versus Rolling Resistance – Mild Wear at Shoulder..................... 71
Table 36. Projected Lifetime Versus Rolling Resistance – Severe Wear at Shoulder ................. 72
Table 37. Analysis of Rolling Resistance Change........................................................................ 76
vii
LIST OF EQUATIONS
Equation 1. Rolling Resistance Calculation, Force Method (ISO 28580)...................................... 9
Equation 2. Rolling Resistance Calculation, Torque Method (ISO 28580) ................................. 10
Equation 3. Input Cycle ................................................................................................................ 27
Equation 4. SAE J1269 Linear Regression Equation for Passenger Car Tires............................. 87
Equation 5. ISO 28580 Rolling Resistance Coefficient................................................................ 89
Equation 6. T&RA Load Formula for “P” Type Tires (S.I. Units) ............................................ 100
viii
DEFINITIONS
SAE – The Society of Automotive Engineers International is an international standards organiza

tion providing voluntary standards to advance the state of technical and engineering sciences.
SAE International, 400 Commonwealth Drive, Warrendale, PA 15096-0001, Tel 877-606-7323,
www.sae.org
ISO – The International Organization for Standardization is a worldwide federation of national

standards bodies that prepares standards through technical committees comprised of international
organizations, governmental and non-governmental, in liaison with ISO. ISO Central Secretariat,
1, ch. de la Voie-Creuse, Case postale 56, CH-1211 Geneva 20, Switzerland, Telephone +41 22
749 01 11, Fax +41 22 733 34 30, www.iso.org
SAE J1269 (Rev. September 2006) – “SAE multi-point standard: Rolling Resistance Measure
ment Procedure for Passenger Car, Light Truck and Highway Truck and Bus Tires: This proce
dure is intended to provide a standard method for gathering data on a uniform basis, to be used
for various purposes (for example, tire comparisons, determination of load or pressure effects,
correlation with test results from fuel consumption tests, etc.).” A single-point test condition
(SRC or standard reference condition) is included. The rolling resistance at this condition may be
calculated from regression of the multi-point measurements or measured directly at the SRC.
SAE J2452 (Issued June 1999) – “Stepwise Coastdown Methodology for Measuring Tire Roll
ing Resistance: This SAE Recommended Practice establishes a laboratory method for determina
tion of tire rolling resistance of Passenger Car and Light Truck tires. The method provides a stan
dard for collection and analysis of rolling resistance data with respect to vertical load, inflation
pressure, and velocity. The primary intent is for estimation of the tire rolling resistance contribu
tion to vehicle force applicable to SAE Vehicle Coastdown recommended practices J2263 and
J2264.”
ISO 18164:2005(E) – “Passenger car, truck, bus and motorcycle tires -- Methods of measuring
rolling resistance: This International Standard specifies methods for measuring rolling resistance,
under controlled laboratory conditions, for new pneumatic tyres designed primarily for use on
passenger cars, trucks, buses and motorcycles.”
ISO 28580 Draft International Standard (DIS) – “Tyre Rolling Resistance measurement
method – single-point test and measurement result correlation – designed to facilitate interna
tional cooperation and, possibly, regulation building. Passenger Car, Truck and Bus Tyres: This
recommendation specifies methods for measuring rolling resistance, under controlled laboratory
conditions, for new pneumatic tyres designed primarily for use on passenger cars, trucks and
buses. Tyres intended for temporary use only are not included in this specification. This includes
a method for correlating measurement results to allow inter-laboratory comparisons. Measure
ment of tyres using this method enables comparisons to be made between the rolling resistance
of new test tyres when they are free-rolling straight ahead, in a position perpendicular to the
drum outer surface, and in steady-state conditions.”
ix
Rolling Resistance (Fr) (ISO/DIS 28580) – “Loss of energy (or energy consumed) per unit of
distance travelled. NOTE 1: The SI unit conventionally used for the rolling resistance is the new
ton metre per metre (N m/m). This is equivalent to a drag force in newtons (N).” (Also referred
to as “RRF”).
Rolling Resistance Coefficient (Cr) (ISO/DIS 28580) – “Ratio of the rolling resistance, in new
tons, to the load on the tyre, in knewtons. This quantity is dimensionless.” (Often multiplied by
1000 kg/metric tonne (MT) for reporting. Also referred to as RRC).
Mean Equivalent Rolling Force (MERF) (SAE 2452) – “The average rolling resistance of a
tire, at a given load/inflation condition, over a driving cycle with a specified speed-time profile.
This implicitly weights the rolling resistance for each speed using the length of time spent at that
speed during the cycle.” For the purpose of this document, MERF is a combined weighting of
MERFs calculated using the standard EPA urban and highway driving cycles. Specifically, this
weighting is 55 percent for the EPA Urban (FTP) Cycle and 45 percent for the EPA Highway
Fuel Economy Cycle.
Standard Mean Equivalent Rolling Force (SMERF) (SAE 2452) – “For any tire is the MERF
for that tire under standard load/inflation conditions defined in 3.10.” For this document, the fi
nal SMERF is also calculated by weighting the SMERF obtained for the EPA urban and highway
cycles, as discussed previously for MERF calculation.
Tire Spindle Force, Ft (ISO/DIS 28580) – “Force measured at the tire spindle in newtons.”
Tire Input Torque, Tp (ISO/DIS 28580) – “Torque measured in the input shaft at the drum
axis, measured in newton-meters.”
Capped Inflation (ISO/DIS 28580) – “Inflating the tire and fixing the amount of inflation gas in
the tire. This allows the inflation pressure to build up, as the tire is warmed up while running.”
Parasitic Loss (ISO/DIS 28580) – “Loss of energy (or energy consumed) per unit of distance
excluding internal tire losses, and attributable to aerodynamic loss of the different rotating ele
ments of the test equipment, bearing friction, and other sources of systematic loss which may be
inherent in the measurement.”
Skim Test Reading (ISO/DIS 28580) – “Type of parasitic loss measurement, in which the tire is
kept rolling, without slippage, while reducing the tire load to a level at which energy loss within
the tire itself is virtually zero.”
x
EXECUTIVE SUMMARY
The first phase of development of the tire fuel efficiency rating system consisted of the evalua
tion of five laboratory rolling resistance test methods, using 25 light-vehicle tire models, in du
plicate at two independent laboratories. Results of this evaluation are documented in the Phase 1
report on the project. The agency’s evaluation showed that all of the rolling resistance test meth
ods had very low variability and all methods could be cross-correlated to provide the same in
formation about individual tire types. The rank order grouping of tire types was statistically the
same for each of the rolling resistance test methods evaluated. However, the relative rankings of
the tires within the population of the 25 models tested shifted considerably when tires were
ranked by either rolling resistance force or rolling resistance coefficient.
It was concluded from Phase 1 that while multi-point rolling resistance test methods are neces
sary to characterize the response of a tire’s rolling resistance over a range of loads, pressures,
and/or speeds, either of the two shorter and less expensive single-point test methods were suffi
cient for the purpose of simply assessing and rating individual tires in a common system. Of the
two single-point methods, the ISO 28580 Draft International Standard (DIS) has the advantage of
using defined lab alignment tires to allow comparison of data between labs on a standardized ba
sis. The use of any of the other single or multi-point test standard would require extensive devel
opment of a method to allow direct comparison of results generated in different laboratories, or
even on different machines in the same laboratory. In addition, the Commission of the European
Communities (EU) has selected ISO 28580 international standard as the basis of its rolling resis
tance rating system. Use of ISO 28580 would allow international harmonization of U.S. and
European test practices.
This report summarizes the results of testing done to examine possible correlations between tire
rolling resistance levels and operating parameters such as vehicle fuel economy, wet and dry
traction, and outdoor and indoor treadwear. With the exception of the OE tires on the fuel econ
omy vehicle, all tires used in Phase 2 were previously tested in one to two indoor rolling resis
tance tests in Phase 1. Fifteen different tire models were installed on the same new passenger car
to evaluate the effects of tire rolling resistance levels on vehicle fuel economy using a test that
approximately followed the EPA’s new 5-cyle dynamometer test. A 10-percent decrease in tire
rolling resistance resulted in approximately 1.1-percent increase in fuel economy for the vehicle.
This result was within the range predicted by technical literature. Reducing the inflation pressure
by 25 percent resulted in a small but statistically significant decrease of approximately 0.3 to 0.5
miles per gallon for four of the five fuel economy cycles, excluding the high-speed, high-
acceleration US06 cycle. This value was smaller than many values predicted by technical litera
ture, and possible explanations are being explored.
Sixteen tire models were subjected to dry and wet skid-trailer testing on asphalt and concrete
skid pads. Both the peak (maximum) and slide (fully locked-tire) coefficients of friction were
measured and indexed against the control tire. For the tires studied, there appeared to be no sig
nificant relationship between dry peak or slide numbers and rolling resistance. However, these
tire models exhibited a strong and significant relationship between better rolling resistance and
poorer wet slide numbers. The peak wet slide number displayed the same tendency, but the rela
tionship was much weaker. This may be significant to consumers without anti-lock braking sys
xi
tems (ABS) on their vehicles since the wet slide value relates most closely to locked-wheel
emergency stops. For newer vehicles with ABS or electronic stability control systems, which op
erate in the earlier and higher peak friction range, the tradeoff is less significant. The agency’s
current Uniform Tire Quality Grading Standards (UTQGS) (575.104) rate wet slide traction but
not wet peak traction. For the subset of five tire models subjected to on-vehicle treadwear testing
(UTQGS), no clear relationship was exhibited between tread wear rate and rolling resistance lev
els. For the subset of six tire models subjected to significant amounts of wear in the indoor
treadwear tests, there was a trend toward faster wear for tires with lower rolling resistance.
The Requirements section of the report contains an analysis of the various options in the draft
ISO 28580 rolling resistance test and their likelihood of inducing variability in the test results.
The lab alignment procedure in ISO 28580, which for passenger tires uses two dissimilar tires to
calibrate a test lab to a master lab, states that it will compensate for differences induced from
tests conducted using different options under the test standard. These options include the use of
one of four measurement methods (force, torque, power, or deceleration), textured or smooth
drum surface, correction of data to a 25C reference temperature, and correction of data from
tests conducted on a test drum of less than 2.0-m in diameter to a 2.0-m test drum. The variabil
ity in test results induced by allowing the various test options, as well as the effectiveness of the
temperature and test drum correction equations is not currently known to the agency. Some rec
ommendations are included.
Concluding the report is a special discussion regarding the use of rolling resistance (Fr) or rolling
resistance coefficient (Cr) as the basis for data reporting and ratings. The ISO 28580 standard
calculates a rolling resistance (Fr, energy loss per unit distance) from one of four different meas
urement methods. Since, rolling resistance varies with the load on the tire, and tires of different
load indexes are tested at different loads, the rolling resistance coefficient is used to allow a rela
tive comparison of the energy consumption of tires of all sizes and load ranges. However, the
normalization of Fr to generate Cr is not consistent across the range of tire sizes and load ranges
in what is expected to be about 20,000 different tires in a common system. If the Cr coefficient is
used as a basis, the data will be skewed towards better ratings for larger tires. While this would
have negligible effects for consumers picking out tires of a given size, there are concerns about
the confusion of consumers if the overall tire fuel economy system was to rate tires that consume
more fuel at a given set of conditions better than tires that consume less fuel at those same condi
tions.
xii
1.0 INTRODUCTION
Reducing energy consumption is a national goal for many reasons, from economic and national
security to improving air quality and reducing greenhouse gas emissions. Also, rising energy
prices are having their effect on consumers and businesses, and have contributed to increases in
the Consumer Price Index in recent years. Hall and Moreland define tire rolling resistance “as the
energy consumed per unit distance of travel as a tire rolls under load.”[1] A vehicle’s fuel econ
omy is affected by tire rolling resistance, therefore, fuel saving could be achieved by reducing
tire rolling resistance. Low-rolling-resistance original equipment (OE) tires are used by auto
manufactures to help meet the Federal fuel economy standards for new passenger cars and light
trucks. However, consumers often purchase less fuel-efficient tires when replacing their vehi
cles’ OE tires, as well as when purchasing subsequent sets of replacement tires. For example,
during 2007 there were an estimated 51 million OE passenger and light truck tires sold in the
United States, as opposed to an estimated 237 million replacement passenger and light truck
tires.[2] Therefore, the rolling resistance of replacement tires could have a significant impact on
the fuel economy of the U.S. light-vehicle fleet.
In the Consolidated Appropriations Act of 2004, Congress provided funding through the NHTSA
to the National Academy of Sciences (NAS) 1 to develop and perform a national tire fuel effi
ciency study and literature review.[3] The NAS was to consider the relationship that low rolling
resistance tires designed for use on passenger cars and light trucks have with vehicle fuel con
sumption and tire wear life. The study was to address the potential of securing technically feasi
ble and cost-effective fuel savings from low rolling resistance replacement tires that do not ad
versely affect tire safety, including the impacts on performance and durability, or adversely im
pact tire tread life and scrap tire disposal, and that does fully consider the average American
‘‘drive cycle.’ The study was to further address the cost to the consumer including the additional
cost of replacement tires and any potential fuel savings. The resulting NAS Transportation Re
search Board report of April 2006 concluded that reduction of average rolling resistance of re
placement tires by 10 percent was technically and economically feasible, and that such a reduc
tion would increase the fuel economy of passenger vehicles by 1 to 2 percent, saving about 1 to 2
billion gallons of fuel per year nationwide. However, as is common in such studies, the NAS
committee did not have a mechanism to generate its own test data 2 and conclusions were based
upon available literature and data.[4] The tire industry eventually supplied rolling resistance data
for 214 passenger and light truck tire models to the NAS committee (177 Michelin
1
Ultimately the Committee for the National Tire Efficiency Study of the Transportation Research Board, a division
of the National Research Council that is jointly administered by the National Academy of Sciences, the National
Academy of Engineering, and the Institute of Medicine.
2
NAS cautioned that much of the available technical literature on tire rolling resistance dates back to the mid-1970s
to mid-1980s. Data on “today’s” passenger tires was difficult to obtain.
manufactured, 24 Bridgestone-manufactured, and 13 Goodyear-manufactured passenger and

light truck tires). 3
The Transportation Research Board report suggests that safety consequences of a 10-percent im
provement in tire rolling resistance “were probably undetectable.” However, the committee’s
analysis of grades under UTQGS (FMVSS No. 575.104) for tires in its study indicated that there
was difficulty in achieving the highest wet traction and/or treadwear grades while achieving the
lowest rolling resistance coefficients. This was more noticeable when the sample of tires was
constrained to similar designs (similar speed ratings and diameters). A lack of access to the raw
rating numbers instead of the final grades provided by the manufacturers prohibited a more de
tailed analysis.
Subsequent to the publication of the NAS committee report, NHTSA initiated a research pro
gram to evaluate five laboratory rolling resistance test methods, using 25 currently available light
vehicle tire models, in duplicate at two independent laboratories. Results of this evaluation are
documented in the Phase 1 report of the project. The agency’s evaluation showed that all of the
rolling resistance test methods had very low variability and all methods could be cross-correlated
to provide the same information about individual tire types. Differences of as much as 30 percent
in measured rolling resistance force were observed between different models of tires of the same
size. It was concluded that while multi-point rolling resistance test methods are necessary to
characterize the response of a tire’s rolling resistance over a range of loads, pressures, and/or
speeds, either of the two shorter and less expensive single-point test methods were sufficient for
the purpose of simply assessing and rating individual tires in a common system. Of the two sin
gle-point methods evaluated, the ISO 28580 Draft International Standard (DIS) has the advan
tage of using defined lab alignment tires to allow comparison of data between labs on a standard
ized basis. The use of any of the other single or multi-point test standard would require extensive
development of a method to allow direct comparison of results generated in different laborato
ries, or even on different machines in the same laboratory. Also, the Commission of the Euro
pean Communities (EU) has selected ISO 28580 international standard as the basis of its rolling
resistance rating system. Use of ISO 28580 would allow international harmonization of U.S. and
European test practices.
In December 2007, Congress enacted the Energy Independence and Security Act of 2007 that
mandated that NHTSA establish a national tire fuel efficiency rating system for motor vehicle
replacement tires within 24 months. While the existing research program was sufficient to meet
the requirements for the testing and rating requirements, NHTSA initiated a second phase of re
search to address the safety and consumer information requirements. Portions of Phase 2 of the
project retested up to 15 models of Phase 1 tires, as well the original equipment tires on the fuel
economy test vehicle, to examine possible correlations between tire rolling resistance levels and
operating parameters such as vehicle fuel economy, wet and dry traction, and outdoor and indoor
3
NAS: “Before the committee’s final meeting, several tire manufacturers, acting through the Rubber Manufacturers
Association, made available measurements of the rolling resistance of a sample of more than 150 new replacement
passenger tires as well as some original equipment (OE) tires. Although the sample was not scientifically derived,
the data proved helpful to the committee as it sought to answer the various questions in the study charge. The timing
of the data’s availability late in the study process limited the statistical analyses that could be undertaken by the
committee.” Reference [4], Page ix.
treadwear. This was accomplished through on-vehicle EPA dynamometer fuel economy tests,
wet and dry skid-trailer traction tests, on-vehicle treadwear tests and experimental indoor tread-
wear tests.
1.1 The Concept of Rolling Resistance

In the latest version of the book The Pneumatic Tire, which was commissioned and published by
NHTSA, LaClair describes the concept of rolling resistance in simple terms[5]:
“When a tire rolls on the road, mechanical energy is converted to heat as a result of the
phenomenon referred to as rolling resistance. Effectively, the tire consumes a portion of
the power transmitted to the wheels, thus leaving less energy available for moving the
vehicle forward. Rolling resistance therefore plays an important part in increasing vehicle
fuel consumption. … Rolling resistance includes mechanical energy losses due to aero
dynamic drag associated with rolling, friction between the tire and road and between the
tire and rim, and energy losses taking place within the structure of the tire.”
LaClair also points out that the term rolling resistance is often mistaken as a measure of the force
opposing tire rotation, when instead is actually a measure of rolling energy loss[6]:
“Although several authors recognized the importance of energy consumption, the concept
of rolling resistance as a retarding force has persisted for many years. Schuring provided
the following definition of rolling resistance as a loss in mechanical energy: “Rolling [re
sistance] is the mechanical energy converted into heat by a tire moving for a unit distance
on the roadway.” He proposed the term rolling loss instead of rolling resistance so that
the long-standing idea of a force would be avoided. … Schuring pointed out that although
rolling resistance -- defined as energy per unit distance -- has the same units as force (J/m
= N), it is a scalar quantity with no direction associated with it.”
Defining rolling resistance as an energy loss is advantageous when considering its effects on the
fuel efficiency of a vehicle. The U.S. Department of Energy estimates that approximately 4.2
percent of the total energy available in the fuel you put in your tank is lost to rolling resistance
during the operation of the vehicle (Figure 1).[7] However, Duleep and NAS point out that the
peak first law (thermodynamic) efficiency of a modern spark-ignited gasoline engine is in the 34
36 percent range (40-42% for diesels), and therefore tire rolling resistance consumes about a
third of the usable energy actually transmitted to the wheels (i.e., 1/3 of the available tractive
energy). Therefore, considering rolling resistance in terms of the energy in the fuel tank is not a
useful measure.[8],[9] For instance, in Figure 1 only 12.6 percent of the energy in the fuel is fi
nally transmitted to the wheels. The 4.2 percent of original fuel energy used by rolling resistance
is actually 33 percent (4.2%/12.6%) of the total usable energy available to the wheels.
Only about 15 percent of the energy from the fuel you put in your tank gets used to move your car down
the road or run useful accessories, such as air conditioning. The rest of the energy is lost to engine and
driveline inefficiencies and idling. Therefore, the potential to improve fuel efficiency with advanced tech
nologies is enormous.
Rolling Resistance – 4.2 percent

For passenger cars, a 5 to 7 percent reduction in rolling resistance increases fuel efficiency by 1 percent.
However, these improvements must be balanced against traction, durability, and noise.
Figure from Department of Energy, 2009
Figure 1. Where Does the Energy Go?
Additionally, the contribution of tire rolling resistance to fuel economy varies with the speed of
the vehicle. At lower speeds, tire rolling resistance represents a larger percentage of the fuel con
sumption (Figure 2) than at higher speeds.[10]
Figure 2. Contribution of Tire Rolling Resistance to Vehicle Fuel Economy Versus Speed
(Reprinted with permission from the Automotive Chassis: Engineering Principles,
2nd Edition, Reed Educational and Professional Publishing Ltd., 2001)
In any discussion of rolling resistance, it is important to consider that the rolling resistance level
of a tire evolves during use. It is reported in literature that a tire’s rolling resistance level, and
therefore its effects on vehicle fuel economy, can decrease by more than 20 percent from a new
tread to completely worn.[11],[12] Therefore, calculations of the benefits of lower tire rolling
resistance derived from measurements of new tires will likely understate the benefits to a vehicle
in terms of absolute fuel economy over the lifetime of the set of tires. However, since both new-
vehicle fuel economy and new-tire rolling resistance change with time, and are dependent on us
age conditions, age, and maintenance levels, attempts to calculate lifetime benefit can vary
widely.
While the hysteretic losses of the tire (primarily the tread) consume a large amount of the avail
able tractive energy, the tires also provide the traction necessary to start, stop, and steer the vehi
cles. Substances soft enough to provide traction on wet, dry, snow, dirt, gravel, etc., surfaces will
also wear. Therefore, the topics of rolling resistance, traction, and treadwear are linked in what
the tire industry refers to as the “magic triangle” (Figure 3). The triangle is a useful graphic since
it conveys the point that a shift to improve properties in one corner of the triangle can diminish
properties in both of the other corners if more advanced and often more expensive tire com
pounding and construction technologies are not employed.
Traction
Rolling
Treadwear Resistance
Figure 3. Magic Triangle: Traction, Treadwear, and Rolling Resistance
From a safety standpoint, the obvious concern from the magic triangle is a loss of tire traction to
achieve lower rolling resistance (better vehicle fuel economy). Since 85 percent of all crashes in
2005 occurred during normal dry weather conditions, and 10 percent in the rain (Table 1.), the
effects of lower rolling resistance on wet and dry traction are of primary importance.[13] Longi
tudinal wet and dry tire traction are easily measured with skid-trailer testing. Conversely, while
crashes occur on snow, sleet, and ice about 4 percent of the time, measuring tire traction on the
varying permutations of these surfaces is not easily done.
Table 1. 2005 Motor Vehicle Crash Data From FARS and GES, Crashes by Weather
Condition
Weather Condition All Crashes Percent

Normal (dry) 5,239,000 85.1%
Rain 584,000 9.5%
Snow/Sleet 264,000 4.3%
Other 72,000 1.2%
Total 6,159,000 100%
2.0 METHODOLOGY
2.1 Test Tires

The majority of the tire models selected for Phase 1 were size P225/60R16 or 225/60R16, which
in 2007 was the most popular size of replacement tire in the United States. Phase 1 of the project
evaluated the rolling resistance of 25 passenger and light-truck tire models. However, time and
budget constraints, as well as equipment limitations, limited Phase 2 to retests of 5 to 16 of the
Phase 1 models in different portions of the project (Table 2). The original equipment tires on the
fuel economy test vehicle added a 17th tire model to the Phase 2 test matrix. The Phase 2 tire
models ranged from 14- to 17-inch rim codes, Q to W speed ratings, 9 to 15 lbf (7 to 11 Cr) in
rolling resistance per ISO 28580, 19 to 36 lbs in weight, 300 to 700 in treadwear rating, and A to
AA in UTQGS traction (wet) rating.
The Phase 1 passenger tires, all purchased as new, were not subjected to optional break-ins listed
in the various rolling resistance tests prior to the warm-up and measurement phases of the tests.
Therefore, Phase 1 tires experienced approximately 50 to 75 miles of straight-line mileage on the
laboratory rolling resistance machine prior to Phase 2 testing. This produced no detectable
treadwear, but did serve to break-in the tires. It has been reported by LaClair that tire rolling re
sistance will decrease about 2-5 percent during a break-in period of 60 minutes at 80 km/h (50
total miles).[14] Therefore, it is anticipated that the rolling resistance of the tires retested in
Phase 2 for on-vehicle fuel economy, traction, and treadwear is approximately 2-5 percent lower
than a brand new tire subjected to these tests. However, it should also be noted that most of these
tests are normally completed with tires that are broken-in prior to testing (vehicle fuel economy -
2,000 miles, outdoor traction - 200 miles, outdoor treadwear - 800 miles).
2.1.1 ASTM F2493 Radial Standard Reference Test Tire

Tire model M14 is an ASTM F2493 SRTT tire. The ASTM F2493 - Standard Specification for
P225/60R16 97S Radial Standard Reference Test Tire (SRTT) provides specifications for a tire
“for use as a reference tire for braking traction, snow traction, and wear performance evaluations,
but may also be used for other evaluations, such as pavement roughness, noise, or other tests that
require a reference tire.” The standard contains detailed specifications for the design, allowable
dimensions, and storage of the tires. The F2493 SRTT is a variant of a modern 16-inch Uniroyal
TigerPaw radial passenger vehicle tire and comes marked with a full USDOT Tire Identification
Number and UTQGS grades. The SRTTs were used extensively throughout the laboratory, test
surface, and fuel economy phases of the test program to monitor the stability of the testing. The
SRTTs had the added advantage of being near the center of the range of passenger tire rolling
resistances in the program (Table 2).
Table 2. Phase 2 Tire Models
ISO 28580 Rolling

ISO 28580 Rolling
Tire Model Code
Resistance Coef
UTQGS Tread-
Resistance, Fr
UTQGS Temp.
Speed Rating
UTQGS Trac.
Performance
Load Index
Weight (lbs.)
Model
Level
wear
MFG
Size
(lbf)
ficient , Cr
G12 Goodyear P225/60R16 97 S Integrity 460 A B Passenger All Sea 9.47 7.36 22.0
son, TPC 1298MS
G8 Goodyear 225/60R16 98 S Integrity 460 A B Passenger All Sea 9.83 7.44 22.9
son
son
B11 Bridgestone P225/60R16 97 H Potenza RE92 340 A A High Performance 10.13 7.87 25.1
OWL All Season
son
M14 Uniroyal P225/60R16 97 S ASTM 16" 540 A B ASTM F 2493-06 11.96 9.30 25.5
SRTT Reference
M13 Michelin 225/60R16 98 H Pilot MXM4 300 A A Grand Touring All 12.07 9.13 24.7
Season
son
B10 Bridgestone 225/60R16 98 Q Blizzak - Performance Winter 12.11 9.16 26.9
REVO1*
D10 Cooper 225/60R16 98 H Lifeliner Tour 420 A A Standard Touring All 13.56 10.26 25.2
ing SLE Season
B14 Bridgestone P225/60R16 97 V Turanza LS-V 400 AA A Grand Touring All 13.90 10.80 28.6
Season
U3 Dunlop P225/60R17 98 T SP Sport 4000 360 A B Run Flat 13.91 10.52 36.4
(Sumitomo) DSST
B15 Dayton 225/60R16 98 S Winterforce* - Performance Winter 13.99 10.58 26.7
P5 Pep Boys P225/60R16 97 H Touring HR 420 A A Passenger All Sea 14.02 10.89 25.7
(Cooper) son
R4 Pirelli 225/60R16 98 H P6 Four Sea 400 A A Passenger All Sea 14.98 11.33 24.3
sons son
B13 Bridgestone P225/60R16 97 T Turanza LS-T 700 A B Standard Touring All 15.01 11.66 29.4
Season
B12 Bridgestone P225/60R16 98 W Potenza RE750 340 AA A Ultra High Perform 15.22 11.51 27.4
ance Summer
Original equipment tires on the fuel economy test vehicle.
Standard reference test tires used as control tires throughout all phases of the study.
*Snow tires will not be rated in the national tire fuel efficiency consumer information program.
2.2 Tire Rolling Resistance Test Procedures

Tire rolling resistance is measured in a laboratory under controlled conditions. The test condi
tions vary between the various SAE and ISO test standards, but the basic premise is the same in
that a tire is mounted on a free-rolling spindle with no camber or slip angle, loaded against a
large-diameter powered test drum, turned by the drum to simulate on-road rolling operation, and
some measure of rolling loss evaluated. Referring back to the book The Pneumatic Tire[5]:
“Rolling resistance is the effort required to keep a given tire rolling. Its magnitude de
pends on the tire used, the nature of the surface on which it rolls, and the operating condi
tions - inflation pressure, load and speed.”
This description is important because it emphasizes that rolling resistance is not an intrinsic
property of the tire, rather a function of many operating variables. This is why multi-point labo
ratory tests measure a tire’s rolling resistance over a range of inflation pressures, loads, and for
some tests, a range of speeds. Conversely, single-point point rolling resistance test methods use a
single set of these variables to estimate the rolling resistance of the tire under nominal, straight-
line, steady state operating conditions (the vast majority of a tire’s rolling operation). In the case
of a laboratory test, rolling resistance (energy loss) is calculated by measuring the amount of ad
ditional force, torque, or power necessary to keep the tire rolling at the test conditions. A fourth
method, which is not widely used, is a deceleration method in which the energy source is de-
coupled from the system and the rate of loss of angular momentum (energy loss) imparted by the
tire is measured.
The two domestic test labs used by the agency had machines that used either the force or the
torque measurement method. A picture of a laboratory rolling resistance test using a force
method can be seen in Figure 4. The machine measures a reaction force at the axle of the test tire
& wheel assembly. The drum is brought up to speed and the tire is warmed up to an equilibrium
temperature. The tire is then lightly loaded to measure “parasitic” losses caused by the tire spin
dle friction, aerodynamic losses, and the test drum/drive system bearings. The tire is then loaded
to the test load and successive readings are taken until consistent force values are obtained. Dur
ing the test, the loaded radius (rL) of the tire is measured during the steady-state conditions. In
ISO 28580 the Rolling Resistance (Fr) at the tire/drum interface is calculated from the measured
force at the spindle (Ft), multiplied by a ratio of the loaded tire radius (rL) to the test wheel radius
(R), minus the skim load (Fpl).
Fr = Ft[1+(rL/R)]-Fpl
Equation 1. Rolling Resistance Calculation, Force Method (ISO 28580)
Fr = Calculated Rolling Resis

tance at Tire/Drum Interface
Ft = Spindle
Force

rL
1.7 meter Drum
Figure 4. Force Method Rolling Resistance Test Machine
Another test lab used by the agency used a torque method machine. The torque method measures
the torque required to maintain the rotation of the drum. The drum is connected to the motor
through a “torque cell” (Figure 5). The drum is brought up to speed and the tire is warmed up to
an equilibrium temperature. The tire is then lightly loaded to measure the losses caused by the
axle holding the tire and aerodynamic losses from the tire spinning. The tire is then loaded to the
test load and successive readings are taken until consistent torque (Tt) values are obtained.
Fr = Tt/R-Fpl
Equation 2. Rolling Resistance Calculation, Torque Method (ISO 28580)
80 grit Surface
Torque Cell 1.7 meter

roadwheel
T = torque
Motor
Figure 5. Torque Method Rolling Resistance Test Machine
10
In one additional calculation, the rolling resistance force (Fr) calculated by any of the methods is
divided by the nominal test load on the tire to produce the rolling resistance coefficient (Cr).
Since the rolling resistance coefficient (Cr) is not linear between tires of different load ranges,
the rolling resistance (Fr) for each tire was compared to the traction, treadwear, and fuel econ
omy measures in the Phase 2 analysis.
Tires in Phases 1 and 2 were subjected to up to three tests. The first and possibly second test may
have been the same indoor rolling resistance test or two different tests, followed by traction,
treadwear or fuel economy testing. A detailed test matrix is provided in Appendix 2. A descrip
tion of the laboratory rolling resistance tests used in Phase 1 follows:
2.2.1 ISO Draft International Standard 28580 Single-Point Rolling Resistance

Tires from all 17 tire models used in Phase 2, though not necessarily the exact tires, were previ
ously tested using the draft ISO 28580 test method.
2.2.2 SAE J1269 & ISO 18164 Multi-Point Rolling Resistance

Tires from all 17 tire models in Phase 2, though not necessarily the exact tires, were previously
tested with SAE J1269, and 11 models were previously tested with ISO 18164 (both tests are
very similar). Data from this multi-point test allows estimation of tire rolling resistance at the test
vehicle load and the two inflation pressures used in the vehicle fuel economy testing.
2.2.3 SAE J2452 Multi-Point (Speed Coast Down) Rolling Resistance

With the exception of the original equipment (OE) tires, tires from 16 tire models in Phase 2,
though not necessarily the exact tires, were previously tested with SAE J2452. Data from this
multi-point test allows estimation of tire rolling resistance at the test vehicle load, two inflation
pressures, and speeds used in the vehicle fuel economy testing.
2.3 Fuel Economy Test Vehicle

A 2008 Chevrolet Impala LS was selected as the test vehicle for fuel economy testing since it
came equipped with P225/60R16 tires, and GM original equipment tires have a Tire Performance
Code (TPC) that allows purchase of replacement tires with the same specifications as the OE
tires. These OE tires (tire type G12) became the 17th group of tires in Phase 2 and had the lowest
rolling resistance of any tire tested in the program (Table 2).
2.4 Test Wheels

Tires were tested on wheels of the corresponding “measuring rim width” for their size. Wheels of
each size used in the test program were purchased new, in identical lots to minimize wheel-to
wheel variation. A tire participating in multiple tests throughout the test program was mounted
11
once on a single new wheel and continued to be tested on that same wheel until completion of all
tests.
2.5 Test Matrix

The EISA legislation requires a national tire fuel efficiency consumer information program “to
educate consumers about the effect of tires on automobile fuel efficiency, safety, and durabil-
ity.”[15] Phase 2 of the project was therefore designed to examine the effects of tire rolling resis
tance levels on vehicle fuel economy, traction, and treadwear. Phase 1 tires were retested in one
of five Phase 2 test protocols: On-vehicle EPA dynamometer fuel economy (Dyno. FE), wet and
dry skid-trailer traction, on-vehicle treadwear, an experimental indoor treadwear test, or tread
rubber analysis by thermogravimetric analysis (TGA) and dynamic mechanical analysis (DMA)
(Table 3). Due to time and cost considerations, as well as the physical constraints the fuel econ
omy test vehicle and skid-trailer, the four tests used a subset of the 17 available Phase 2 tire
models selected to cover the range of rolling resistance values in the experiment.
12
Table 3. Test Matrix

Code MFG Size Load Speed Model RR Dyno. Wet & On- Indoor TGA
Index Rating (lbf) FE Dry vehicle Treadwear /
Traction Treadwear DMA
G12 Goodyear P225/60R16 97 S Integrity 9.47 x x
G8 Goodyear 225/60R16 98 S Integrity 9.83 x x x x x
G11 Goodyear P225/60R17 98 S Integrity 10.02 x x x
B11 Bridgestone P225/60R16 97 H Potenza 10.13 x x x x x
RE92 OWL
M14 Uniroyal P225/60R16 97 S ASTM 16" 11.96 x x x x x
SRTT
M13 Michelin 225/60R16 98 H Pilot MXM4 12.07 x x x x x
B10 Bridgestone 225/60R16 98 Q Blizzak 12.11 x x x
REVO1
D10 Cooper 225/60R16 98 H Lifeliner 13.56 x x x
Touring
SLE
B14 Bridgestone P225/60R16 97 V Turanza 13.90 x x x
LS-V
U3 Dunlop P225/60R17 98 T SP Sport 13.91 x x x
(Sumitomo) 4000
DSST
B15 Dayton 225/60R16 98 S Winterforce 13.99 x x x
P5 Pep Boys P225/60R16 97 H Touring HR 14.02 x x x
(Cooper)
R4 Pirelli 225/60R16 98 H P6 Four 14.98 x x x
Seasons
B13 Bridgestone P225/60R16 97 T Turanza 15.01 x x x x x
LS-T
B12 Bridgestone P225/60R16 98 W Potenza 15.22 x x x
RE750
2.6 Tread Compound Properties Testing

The tread rubber of 16 Phase 1 passenger tires was analyzed for compound composition by
thermogravimetric analysis (TGA). The mechanical properties of the treads were evaluated by
dynamic mechanical analysis (DMA). TGA is a useful tool for characterizing polymer composi
tions. The weight loss as a function of temperature has been used to determine polymer loading,
rubber chemical loading, carbon black loading, and ash levels. For polymers with very different
thermal stabilities, the TGA curves can be used to determine the amount of each polymer pre
sent. Thermogravimetric analysis was performed using about 10 mg of sample of each tire tread.
The purge (He) gas flow rate to the TGA was set at 10ml/min during weight loss measurements.
The heating rate was 10C/min to improve the resolution of small variations in the decomposi
tion curves. At 600C, the purge gas was switched over to air for carbon black combustion.
These average values represent the average of three measurements. Figure 6 shows a representa
13
tive weight loss curve with the regions that represent each component identified. The results of
the TGA analysis are shown in Table 4.
120
100
Volatile Components
Weight Retained (%)
80
60
Polymer
40
20 Carbon Black
Ash (Zinc Oxide, Silica, …

0
0 200 400 600 800 1000
Temperataure (degC)
Figure 6. Sample TGA Weight Loss Curve
Table 4. Analysis of Tread Composition by TGA

Tire Black,
Type Volatiles, phr Total Total
Tire Polymer,% phr (25 (550 Ash, phr Filler, Silica, Formulation,
# (325-550C) 325°C) 850C) (Residue) phr phr phr
B10 3104 57 18 32 25 51 19 169
B11 3129 56.8 18 31 27 52 21 170
B12 3154 49 25 54 25 73 19 198
B13 3179 51.3 22 44 29 67 23 189
B14 3204 52 25 13 54 62 48 186
D10 3313 46.9 33 77 3 77 0 207
B15 3337 54.3 19 63 3 63 0 178
U3 3362 52.4 18 33 40 67 34 185
G8 3412 60.4 15 38 12 45 6 159
G9 3441 52.9 23 60 6 60 0 183
G10 3466 58.3 22 45 4 45 0 165
G11 3491 63.3 15 33 11 37 5 152
M13 3620 54.3 19 10 55 59 49 178
14
Tire Black,
Type Volatiles, phr Total Total
Tire Polymer,% phr (25- (550- Ash, phr Filler, Silica, Formulation,
# (325-550C) 325°C) 850C) (Residue) phr phr phr
P5 3670 47.1 29 79 4 79 0 206
R4 3695 48.3 30 42 35 71 29 201
M14 3720 55 19 30 32 57 26 176
Typical examples of temperature sweep data by the tension method and the shear method are
shown below in Figure 7 and Figure 8. The viscoelastic (dynamic mechanical) properties of a tire
tread have been correlated to the performance of tires.[16],[17],[18],[19] Decreased tangent  at
60C is used as a predictor of the tread compound’s contribution to tire rolling resistance. In-
creased tangent  at 0C has been shown to correlate to the wet traction performance of the tire.
Since these properties tend to move in parallel, lowering the tangent  at 60C while maintaining
a high tangent  at 0C normally requires utilization of advanced and often more expensive com-
pounding technologies. The DMA results for high tangent  at 0C and 60C are shown in Table
5.
0.8
0.7
0.6
Tangent Delta
0.5
0.4
0.3
0.2
0.1
0.0
-150 -100 -50 0 50 100
Temperatue (C)
Figure 7. Tan  as a Function of Temperature From the Tension Test
15
0.6
0.5
0.4
Tangent Delta
0.3
0.2
0.1
0
-100 -50 0 50 100
-0.1
Temperature (deg C)
Figure 8. Tan  as a Function of Temperature From the Shear Test
Table 5. DMA Results for Tangent  at 0C and 60C
Tire Tire Rolling Tension Shear

Type # Resistance* Tan at Tan at Ratio 0/60 Tan at Tan at Ratio 0/60
(lbf) 0°C 60°C 0°C 60°C
G8 3412 9.83 0.169 0.0762 2.22 0.164 0.0689 2.38
G11 3491 10.02 0.174 0.086 2.02 0.177 0.0754 2.35
B11 3129 10.13 0.194 0.0771 2.52 0.174 0.067 2.60
G9 3441 11.26 0.245 0.188 1.30 0.18 0.152 1.18
M14 3720 11.96 0.287 0.193 1.49 0.202 0.146 1.38
M13 3620 12.06 0.254 0.147 1.73 0.168 0.117 1.44
G10 3466 12.09 0.242 0.181 1.34 0.184 0.151 1.22
B10 3104 12.11 0.2 0.155 1.29 0.16 0.133 1.20
D10 3313 13.56 0.26 0.192 1.35 0.183 0.16 1.14
B14 3204 13.90 0.313 0.145 2.16 0.233 0.132 1.77
U3 3362 13.91 0.256 0.173 1.48 0.202 0.147 1.37
B15 3337 13.98 0.208 0.15 1.39 0.158 0.123 1.28
P5 3670 14.02 0.271 0.207 1.31 0.161 0.156 1.03
R4 3695 14.98 0.296 0.201 1.47 0.211 0.159 1.33
B13 3179 15.01 0.265 0.168 1.58 0.19 0.138 1.38
B12 3154 15.22 0.387 0.193 2.01 0.28 0.146 1.92
*ISO 28580 single-point rolling resistance
2.7 On-Vehicle Fuel Economy Testing

The effects of tire rolling resistance on automobile fuel efficiency was evaluated by installing 15
different tire models on a new 2008 Chevrolet Impala LS and evaluating its fuel economy in the
2008 five-cycle EPA fuel economy test.[20] Testing was completed under contract by the Trans
portation Research Center, Inc. (TRC, Inc.) emissions laboratory. Since tire inflation pressure
affects the operational rolling resistance of a tire, the vehicle fuel economy measurements were
conducted at two different tire inflation pressures. Testing was completed at the vehicle placard
16
pressure of 210 kPa (30 psi). Six models were tested at both the placard inflation pressure of 210
kPa and at 158 kPa (23 psi), which represents the tire pressure monitoring system (TPMS) acti
vation threshold of 25 percent inflation pressure reduction. It is important to note, for reasons
that will be explained, that these tests were research and not official EPA fuel economy ratings
of the test vehicle. The many tire sets and repeats of test for statistical analysis/dual inflation
pressure resulted in the test vehicle acquiring nearly 6,000 miles by the end of testing. The EPA
estimates that new vehicles will not obtain their optimal fuel economy until the engine has bro
ken in at around 3,000 to 5,000 miles.[21] Therefore the fuel economy of the test vehicle was
expected to improve slightly during the course of testing, a factor that was tracked and accounted
for by the repeated testing of the control and OE tires at regular intervals throughout the testing.
2.7.1 EPA 40 CFR Part 86 Dynamometer Fuel Economy Testing

Per EPA 40 CFR Part 86, the new 2008 Chevrolet Impala LS test vehicle was broken in for
2,000 miles on a test track. To keep the original equipment tires in the same low mileage state as
the Phase 1 tires, the vehicle was broken-in on a spare set of replacement tires of the original
equipment size. For this reason, even the fuel economy tests of the Impala with the original
equipment tires were not official EPA test numbers. The original equipment tires were re
installed on the vehicle at placard inflation pressure and the road load coastdown procedure was
completed. The coastdown procedure generates vehicle-specific coefficients for dynamometer
settings and fuel economy calculations.
The fuel economy dynamometer is housed in an environmental chamber to control the tempera
ture for ambient (68 to 86 degrees F), heated (95 degrees F) or cold (20 degrees F) temperatures.
The vehicle dynamometer is a 1.22-meter (48-inch) diameter, smooth surface drum located in the
floor of the chamber. The vehicle is placed atop the dynamometer rolls and restrained to prevent
movement (Figure 9a). A fan meeting standard specifications is located in front of the vehicle to
provide cooling (Figure 9b). A computer is mounted inside the vehicle to provide the driver with
a prescribed speed pattern that must be followed for each test cycle (Figure 9c). The exhaust gas
is routed from the vehicle exhaust tailpipe via hoses to a collection system connected to gas ana
lyzers (Figure 9d).
17
Figure 9a. Tire on 1.22 Meter Dynamometer Figure 9b. Chamber and Fan
Figure 9c. Drive Cycle Computer Figure 9d. Exhaust Coupling

Figure 9. Vehicle Fuel Economy Dynamometer Testing
Details of the 2008 EPA fuel economy test can be found in Table 6, which is from the EPA’s
www.fueleconomy.gov Website.[22]
18
Table 6. 2008 EPA Fuel Economy 5-Driving Schedule Test (Source: EPA, 2009)
Driving Test Schedule
Schedule
Attributes City (FTP) Highway High Speed AC (SC03) Cold
(HwFET) (US06) Temp (Cold CO)
Trip Type Low speeds in Free-flow traffic Higher speeds; AC use under City test w/ colder
stop-and-go at highway harder acceleration hot ambient outside tempera
urban traffic speeds & braking conditions ture
Top Speed 56 mph 60 mph 80 mph 54.8 mph 56 mph
Average 21.2 mph 48.3 mph 48.4 mph 21.2 mph 21.2 mph
Speed
Max. Accel 3.3 mph/sec 3.2 mph/sec 8.46 mph/sec 5.1 mph/sec 3.3 mph/sec
eration
Simulated 11 mi. 10.3 mi. 8 mi. 3.6 mi. 11 mi.
Distance
Time 31.2 min. 12.75 min. 9.9 min. 9.9 min. 31.2 min.
Stops 23 None 4 5 23
Idling time 18% of time None 7% of time 19% of time 18% of time
Engine Cold Warm Warm Warm Cold
Startup*
Lab tem 68-86ºF 95ºF 20ºF
perature
Vehicle air Off Off Off On Off
condition
ing
*A vehicle's engine doesn't reach maximum fuel efficiency until it is warm.
Whole vehicle preconditioning must be done between the ambient and cold test cycles. There
fore, instead of running all five fuel economy cycles sequentially in their traditional order, testing
with the 15 sets of tires was split into blocks that facilitated a much more rapid test throughput.
In addition, to gather more data for statistical purposes, two extra HwFET cycles were run se
quentially after the first HwFET cycle. The testing was conducted at the placard tire inflation
pressure of 210 kPa (30 psi) and repeated at the TPMS warning activation pressure of 158 kPa
(22.3 psi) for selected tires.
Vehicle Preconditioning
Vehicle preconditioning begins with draining the existing fuel from the vehicle’s fuel tank and
replacing it with a 40 percent fuel tank capacity fill of the specified fuel. The vehicle is then
driven through one Urban Dynamometer Driving Schedule (UDDS). This procedure is followed
by a soak period of at least 12 hours, but not exceeding 36 hours. All preconditioning procedures
are performed at the conditions of the test schedule.
FTP Schedule Testing

Following the vehicle’s soak period, the vehicle is pushed, not driven, onto a chassis dynamome
ter for a cold start exhaust emissions test (’75 FTP). The Federal test procedure (FTP) simulates
normal city driving and collects dilute exhaust emissions into bags for analysis in three phases:
the cold transient (CT), the cold stable (CS), and the hot transient (HT). The UDDS is followed
during the CT and CS, and, following a ten-minute soak on the dynamometer, the first phase, or
19
bag, of the UDDS is repeated for the HT. The results of these phases are combined to provide
grams per mile (g/mi) for total hydrocarbons (THC), non-methane hydrocarbons (NMHC), car
bon monoxide (CO), carbon dioxide (CO2) and oxides of nitrogen (NOx). Fuel economy, in miles
per gallon, is determined via the carbon balance method.
HwFET Schedule Testing

Following each FTP test, the vehicle is kept on the chassis dynamometer and the Highway FET
(HwFET) driving cycle was run twice. The first running of the HwFET served only to stabilize
vehicle temperatures and emissions, therefore fuel economy was not measured during this cycle.
The cycle is repeated and all emissions measurements are taken as described for FTP testing with
the exception that a single bag is used to collect the dilute exhaust sample (single phase). Fuel
economy, in miles per gallon, is again determined via the carbon balance method. The Phase 2
testing protocol added two additional repeats for the HwFET cycle that were run and measured
sequentially.
US06 Schedule Testing

This test type is the aggressive-driving portion of the supplemental FTP (SFTP), consisting of
higher speeds and acceleration rates.
SC03 (AC2 Alternate) Schedule Testing

This test type has been introduced to represent the engine load and emissions associated with the
use of air conditioning units in vehicles. Since the TRC, Inc. emissions lab lacks the solar-
loading equipment necessary to run a full SC03 test, the AC2 alternative was used. This alterna
tive was only valid for 2000-2001 model year vehicles unless approved by the EPA, therefore the
result for each individual cycle is reported in this report but not composite 5-cycle numbers for
the vehicle.[23] The AC2 alternative mimics the SC03 except that the thermal load is simulated
by placing the vehicle’s air conditioning temperature control to full hot, air conditioning on, and
the driver’s side window left down. In addition, the test cell is kept at 76 F and 50 grains of wa
ter per pound of dry air versus the SC03 requirement of 95 F and 100 grains of water per pound
of dry air. All other procedures follow the SC03.
Cold CO Schedule Testing

This test follows the same driving cycle as the FTP, but the test is performed at 20 F and the
vehicle is filled with Cold CO specific fuel. The vehicle is operated through one UDDS
preparation cycle at 20° F. Then, the vehicle is parked in a soak chamber maintained at 20 F for
a minimum of 12 and a maximum of 36 hours prior to beginning each test. Following the 20 F.
soak, the vehicle is pushed into the dynamometer chamber (which is at 20 F) and then operated
through the normal FTP test.
The program was completed in blocks of tests, with the M14 control tires and G12 OE tire run
multiple times to track possible vehicle, tire and test equipment drift. The completed test cycles
are summarized in Table 7.
20
Table 7. Fuel Economy Test Schedules

Pressure City (FTP) Highway (HwFET)* High Speed (US06) AC (SC03) Cold
Temp (Cold CO)
210 kPa 19 57 19 19 19
158 kPa 6 16 6 6 6
*Two extra cycles completed after first run to gauge statistical variability.
2.8 Skid-Trailer Tire Traction Testing

FMVSS No. 575.104, Uniform tire quality grading standards requires manufacturers to provide
a (wet slide) traction grade for all tires subject to standard and manufactured after April 1, 1980.
A formal description follows[24]:
“To assist consumers purchasing new vehicles or replacement tires, NHTSA has rated
more than 2,400 lines of tires, including most used on passenger cars, minivans, SUVs
and light pickup trucks. Traction grades are an indication of a tire's ability to stop on wet
pavement. A higher graded tire should allow a car to stop on wet roads in a shorter dis
tance than a tire with a lower grade. Traction is graded from highest to lowest as "AA",
"A", "B", and "C". Of current tires: 3 percent are rated “AA”, 75 percent are rated “A”,
22 percent are rated “B”, only 1 line of tires rated “C”.”
The UTQGS skid-trailer traction testing was performed at the NHTSA test facility on Goodfel
low Air Force Base in San Angelo, Texas. The traction grading tests are now performed on a
purpose-built oval at the base rather than the original test surface diagram shown in 575.104. The
test pavements are asphalt and concrete skid pads constructed in accordance with industry speci
fications for skid surfaces. ASTM E 501 4 reference (control) tires are used to monitor the trac
tion coefficient of the two surfaces (which varies based on environmental conditions, surface
wear, etc.). During a normal wet traction test, a vehicle tows a skid-trailer (Figure 10) at 40 mph
across the test surfaces. Water is dispersed ahead of the tire from a water nozzle just before the
brake is applied. Instrumentation measures the horizontal force as the brake is applied to one
wheel of the trailer until lock-up, and then held for a few seconds and released. The tests are re
peated for a total of 10 measurements on each surface. The candidate (test) tires are conditioned
by running for 200 miles on a pavement surface. The candidate tires are then fitted to the trailer,
loaded to a specified load and pressure, then subjected to the same testing completed on the con
trol tires. The average sliding coefficient of friction for the candidate tire on each surface is cor
rected using the coefficients of the control tire to yield an adjusted traction coefficient for the
candidate tire on each test surface.
4
ASTM E 501-94 Standard Specification for Standard Rib tire for Pavement Skid Resistance Tests. Available from
American Society for Testing and Materials, http://astm.org.
21
Figure 10. NHTSA San Angelo Skid-Trailer
Phase 2 traction tests were conducted with tires of 16 models previous tested in Phase 1. Two
tires had the highest traction grade “AA,” 14 tires were graded “A” (Table 8). Since these tires
experienced some break-in during the 50- to 70-mile rolling resistance tests, these tires were only
conditioned for 70 miles on a pavement surface rather than the normal 200 miles. 5 Since the tires
were not new, and had a reduced break-in, the results generated are for research purposes and are
unofficial. The test matrix was also repeated on dry asphalt and concrete test surfaces. The num
ber of measurements on the dry surfaces was reduced to preserve the limited test surface area
from rubber buildup.
Since modern antilock brakes (ABS) and electronic stability control (ESC) operate in the lower
slip and higher friction region, the peak coefficient recorded during the traction testing was also
used for comparisons in Phase 2 in addition to the slide values used for UTQGS wet traction.
5
Two additional tires of a Phase 1 tire model were broken -in for the full 200 miles and compared to a set of two
that had the 50- to 70-mile roadwheel break-in. There was no significant difference in their traction numbers.
22
Table 8. Phase 2 Wet and Dry Skid-Trailer Test Tires
ISO 28580 Rolling

Tire Model Code
UTQGS Tread-
Resistance, Fr
UTQGS Temp.
Speed Rating
UTQGS Trac.
Performance
Load Index
Weight (lbs.)
Model
Level
wear
MFG
Size
(lbf)
B14 Bridgestone P225/60R16 97 V Turanza LS-V 400 AA A Grand Touring All Season 13.90 28.6
B12 Bridgestone P225/60R16 98 W Potenza RE750 340 AA A Ultra High Performance Sum 15.22 27.4
mer
D10 Cooper 225/60R16 98 H Lifeliner Touring SLE 420 A A Standard Touring All Season 13.56 25.2
P5 Pep Boys P225/60R16 97 H Touring HR 420 A A Passenger All Season 14.02 25.7
(Cooper)
R4 Pirelli 225/60R16 98 H P6 Four Seasons 400 A A Passenger All Season 14.98 24.3
B11 Bridgestone P225/60R16 97 H Potenza RE92 OWL 340 A A High Performance All Season 10.13 25.1
M13 Michelin 225/60R16 98 H Pilot MXM4 300 A A Grand Touring All Season 12.07 24.7
B13 Bridgestone P225/60R16 97 T Turanza LS-T 700 A B Standard Touring All Season 15.01 29.4
M14 Uniroyal P225/60R16 97 S ASTM 16" SRTT 540 A B ASTM F 2493-06 Reference 11.96 25.5
G8 Goodyear 225/60R16 98 S Integrity 460 A B Passenger All Season 9.83 22.9
G11 Goodyear P225/60R17 98 S Integrity 460 A B Passenger All Season 10.02 24.5
U3 Dunlop P225/60R17 98 T SP Sport 4000 DSST 360 A B Run Flat 13.91 36.4
(Sumitomo)
B10 Bridgestone 225/60R16 98 Q Blizzak REVO1* - Performance Winter 12.11 26.9
B15 Dayton 225/60R16 98 S Winterforce* - Performance Winter 13.99 26.7
*Snow tires will not be rated in the national tire fuel efficiency consumer information program.
2.9 On-Vehicle Tire Treadwear Testing

FMVSS No. 575.104, Uniform tire quality grading standards requires manufacturers to provide
a treadwear grade for all tires subject to standard and manufactured after April 1, 1980. A formal
description follows[25]:
“Treadwear grades are an indication of a tire's relative wear rate. The higher the tread-
wear number is, the longer it should take for the tread to wear down. A control tire is as
signed a grade of 100. Other tires are compared to the control tire. For example, a tire
grade of 200 should wear twice as long as the control tire. Of current tires: 15 percent are
rated below 200, 25 percent are rated 201 - 300, 32 percent are rated 301 - 400, 20 per
cent are rated 401 - 500, 6 percent are rated 501 - 600, 2 percent are rated above 600.”
Additional tires from five of the six models used in UTQG traction testing were tested in the
UTQGS treadwear test. The five tires with treadwear grades ranging from 300 to 700 were
mounted and balanced on 16 x 7.0" rims. The groove depths of the tires were then measured. All
tires were measured with groove one being the outside groove on the serial side. The tires were
23
then installed on five Mercury Marquis vehicles for testing on the UTQG test route near San An
gelo, Texas (Table 9). The vehicles were loaded to 1,182 pounds per wheel within +/-1 percent.
The vehicles were aligned to center of manufacturer's specifications for caster and camber and
toe.
Table 9. On-Vehicle Treadwear Testing
ISO 28580 Rolling

Tire Model Code
UTQGS Tread-
Resistance, Fr
UTQGS Temp.
Speed Rating
UTQGS Trac.
Performance
Load Index
Weight (lbs.)
Model
Level
wear
MFG
Size
(lbf)
The nine-day test conducted consisted of 400-mile day shifts and 400-mile night shifts, for a total
of 7,200 miles including break-in. A Shadow Tracker tracking device was placed in the lead ve
hicle at the beginning of each day shift to record speed, miles traveled, and stops made. The
route is described in Figure 11. The tires were rotated on the vehicle every 400 miles and meas
ured every 800 miles. The vehicles were aligned every 800 miles. The vehicles were rotated
through the convoy at the end of every 800 miles after break-in. The tires from all the vehicles
were rotated from vehicle to vehicle every 1,600 miles after break-in. During the course of the
test, the highest temperature was 93 degrees Fahrenheit and the lowest temperature was 47 de
grees Fahrenheit. The average high for the nine days was 82.6 degrees Fahrenheit and the aver
age low was 63.8 degrees Fahrenheit. There were 581 wet miles during the nine days of testing.
Testing was put on hold for three days due to road closures on the South Loop. The tires were
then measured at the end of the convoy testing to determine the loss of tread depth. More detail
of this test may be found in FMVSS No. 575.104.
24
Figure 11. UTQGS Treadwear Course
2.10 Indoor Tire Treadwear Testing

The FMVSS No. 575.104 requires all passenger tires (with some exceptions) manufactured after
April 1, 1980, to be graded for tread life. However, advances in radial tire tread compounding
since 1980 have resulting in longer life treads that exhibit only a marginal amount of wear after
running the 7,200-mile UTQGS treadwear course. To evaluate the effects of bulk treadwear on
tire rolling resistance, additional tires of the five Phase 1 tire models subjected to on-vehicle
treadwear, as well as original equipment tires from the Impala fuel economy vehicle, were sub
jected to a more aggressive indoor treadwear test developed by Smithers Scientific Services, Inc.
in Ravenna, Ohio (Table 10).
25
Table 10. Indoor Treadwear Testing
ISO 28580 Rolling

Tire Model Code
UTQGS Tread-
Resistance, Fr
UTQGS Temp.
Speed Rating
UTQGS Trac.
Performance
Load Index
Weight (lbs.)
Model
Level
wear
MFG
Size
(lbf)
G12 Goodyear P225/60R16 97 S Integrity 460 A B Passenger All Season, TPC 9.47 22.0
1298MS
The testing was completed on an MTS 860 Durability Machine (Figure 12a), 3.048-meter (120
inch) diameter drum covered with 3M 180µ (microfinishing) film with servo hydraulic control of
tire radial load, tire slip angle and/or slip load, tire camber angle, road way speed, and braking
torque. A powder spray system is used to prevent rubber buildup on the drum 3M surface. The
machine was programmed with a drive file that allows for consistent application of energy. The
machine was run in force control so that the amount of energy input to the tire/wheel assembly
was consistent between test specimens.
Two test methods were conducted: one was a 25 percent Fz (radial load) test and the other was a
20 percent Fz test. Two tires of each of the six tire models were tested using the 25 percent test.
One each of the five Phase 1 tire models were tested using the less demanding 20 percent test.
The tires were of two load indexes and therefore tested using two different load and force levels
to match the rolling resistance load differences. Table 11 lists these test conditions.
Table 11. Test Parameters

Item P225/60R16 97S 225/60R16 98S
Radial Load – 80% Max (lbs / N) 1287 / 5725 1322 / 5882
Camber Angle (  ) 0 0
Speed (mph / km/h) 50 / 80 50 / 80
Inflation Pressure ( psi / kPa) 30.5 / 210 30.5 / 210
Fy (Lateral) Amplitude – 25% (lbs / N) 322 / 1431 331 / 1471
Fy (Lateral) Amplitude – 20% (lbs / N) 257 / 1145 264 / 1176
Recognizing the historical significance of side force, a frictional work or work rate approach was
conducted in which the side force was the controlled parameter and was varied throughout the
wear test.[26] The 25 percent Fz test consisted of 1,641 lateral force cycles. The input cycle was
a sine wave of the following form, where Fz is the radial load and t is the time in seconds:
26
1 
Fy  25%Fz sin   t 
 15 
Equation 3. Input Cycle
A similar cycle was used for the 20 percent Fz profile as well, except the coefficient was equal to
20 percent Fz. Data that was collected as part of the wear testing were tire/wheel assembly
weight, and laser profilometry using a precision scale and a Bytewise CTWIST machine (Figure
12b). The CTWIST machine collects 4,096 data points per tire revolution every millimeter
across the tire. The data was collected at the new or pre-test point, at the halfway point, and at
the end of the test. This allows for wear rate to be evaluated.
The test sequence required the tire wheel assemblies to be weighed, laser-profiled, measured for
rolling resistance using the proposed ISO 28580 single-point test method, and then run on a 400
mile indoor wear cycle. The tires were weighed, laser-profiled, and measured for rolling resis
tance before the final wear cycle of 400 miles was conducted. After the final wear cycle the tires
were then again weighed, laser-profiled, and measured for rolling resistance in their final state.
Figure 12a. MTS 860 Durability Machine Figure 12b. CTWIST Machine
Figure 12. Indoor Treadwear Equipment
27
3.0 RESULTS
3.1 Effect of Tire Rolling Resistance on Automobile Fuel Efficiency

Fifteen tire models with varying rolling resistance levels were tested on a single vehicle for dy
namometer fuel economy as previously described. Six models were tested at both the placard
inflation pressure of 210 kPa and at 158 kPa, which represents the TPMS activation threshold of
25 percent inflation pressure loss. The effects of tire rolling resistance on vehicle fuel economy
are known to be on the scale of fractions of a mile per gallon per pound of rolling resistance.
Therefore, from the outset of the program it was known that the fuel economy tests were at
tempting to measure rolling resistance effects that were at the limits of the accuracy of the test
procedure. In an attempt to account for this, the SRTT (tire type M14) control tires were tested
periodically throughout the testing sequence to monitor possible data shifts. These may be of two
types:
1. Shifts in the data due to an event(s) during the approximately four month test pro
gram, or
2. Drift in the data due to gradual changes in vehicle or dynamometer function.
The data is shown in Table 12.
28
Table 12. Test Matrix by Date
Rolling Re-
City,
Air Condi-
Pressure,
Tire Type
Highway,
Highway,
Highway,
City, FTP
sistance,
Inflation
tioning,
FET (1)
FET (2)
FET (3)
Speed,
SC03
US06
Cold
High
Date
FTP
kPa
lbs
09/30/08 M14 6 210 11.96 21.3 35.9 36.8 36.2 19.3 21.9
10/01/08 B11 210 10.13 21.3 36.7 37.1 36.8 19.6 22.1
10/02/08 B13 210 15.02 20.8 34.8 35.4 34.6 18.6 21.1
10/06/08 B13 158 15.58 7 20.3 34.3 35.2 34.6 18.5 20.9
10/07/08 G8 158 10.45 21.4 36.5 37.2 36.6 19.3 22.2
10/08/08 M13 210 12.06 21.1 35.8 36.3 36.0 19.4 21.9
10/09/08 M14 158 12.47 21.1 35.9 36.5 36.1 19.3 21.9
10/10/08 G8 210 9.83 21.9 37.3 38.2 37.8 19.9 22.4
10/13/08 M13 158 12.60 21.1 35.5 36.4 35.7 19.3 21.9
10/14/08 B11 158 10.80 21.3 36.3 37.1 36.3 19.3 21.9
10/15/08 M14 210 11.96 21.4 36.2 36.7 36.4 19.3 21.7
10/16/08 G12 210 9.47 21.7 37.5 38.2 37.8 20.1 22.4
10/17/08 G12 158 10.09 21.2 36.5 37.3 36.7 19.4 22.3
10/21/08 M14 210 11.96 18.8
10/22/08 M14 158 12.47 18.4
10/23/08 B11 210 10.13 19.0
10/24/08 B11 158 10.80 18.8
10/25/08 B13 158 15.58 18.1
10/27/08 B13 210 15.01 18.4
10/28/08 G8 210 9.83 19.0
10/30/08 M13 158 12.60 18.3
10/31/08 M13 210 12.06 18.7
11/03/08 G8 158 10.45 18.9
11/05/08 G12 210 9.47 19.3
11/06/08 G12 158 10.09 19.1
11/10/08 M14 210 11.96 21.7 36.8 37.3 36.4 20.0 21.5
11/11/08 B10 210 12.11 21.2 35.9 37.2 35.8 19.6 22.0
11/12/08 B12 210 15.22 20.8 35.4 36.1 35.2 19.4 21.8
11/13/08 B14 210 13.90 21.1 35.7 36.8 35.9 19.5 22.0
11/14/08 D10 210 13.56 21.4 36.1 37.1 36.2 19.6 22.0
11/17/08 B15 210 13.99 20.4 35.8 36.4 35.9 19.3 20.8
11/18/08 U3 210 13.91 21.0 36.0 36.8 36.3 19.5 20.6
11/19/08 G11 210 10.02 21.5 36.2 37.1 36.9 20.9 22.2
11/20/08 P5 210 14.02 21.1 35.7 36.7 35.6 21.5 22.0
11/21/08 R4 210 14.98 20.9 35.7 36.4 21.4 22.1
11/24/08 M14 210 11.96 21.7 37.5 36.5 21.9 22.0
11/25/08 G12 158 10.09 Special Tests (Collection Bag Comparison)
12/02/08 M14 210 11.96 19.1
12/03/08 B10 210 12.11 19.1
12/04/08 B12 210 15.22 18.3
12/05/08 B14 210 13.90 18.5
12/06/08 D10 210 13.56 18.7
6
Bold values denote runs with tire types G12 and M14 that were systematically repeated during the testing
7
Rolling resistance values were estimated for 158 kPa tires by adjusting for pressure using regression coefficients
from ISO 18164 and SAE J1269 multi-point testing for the tire type
29
City,
Rolling Re
Air Condi
Pressure,
Tire Type
Highway,
Highway,
Highway,
City, FTP
sistance,
Inflation
tioning,
FET (1)
FET (2)
FET (3)
Speed,
SC03
US06
Cold
High
Date
FTP
kPa
lbs
12/08/08 B15 210 13.99 18.2
12/10/08 U3 210 13.91 18.2
12/11/08 G11 210 10.02 19.0
12/12/08 P5 210 14.02 18.5
12/15/08 R4 210 14.98 18.4
12/16/08 G12 210 9.47 18.6
12/17/08 M14 210 11.96 18.8
01/13/09 M14 210 11.96 21.7 37.2 37.0 37.1
01/14/09 M14 210 11.96 21.7 37.3 37.3 37.5 19.8 21.9
01/15/09 G12 210 9.47 22.1 37.8 37.8 37.9 20.3 22.5
01/19/09 B13 210 15.01 20.6 36.0 35.7 35.9 19.5 20.8
01/21/09 G11 210 10.02 22.1 38.0 38.3 37.9 20.4 22.6
01/22/09 P5 210 14.02 21.4 36.2 36.7 36.9 19.6 20.9
01/23/09 R4 210 14.98 21.2 36.3 36.7 36.5 19.6 20.8
Legend:
Not Scheduled
Mis-Test
3.1.1 Preliminary Analysis: Data Shifts

There were three events identified which corresponded to possible shifts or block events in the
data.
1. The Highway FET fuel economy cycle was run in triplicate, sequentially and showed
systematic differences between the runs.
2. A shift to significantly higher air conditioning SC03 cycle was noted on November
20, 2008. This followed investigation of the differences found in the Highway FET
values.
3. Physical replacement of the valves and recalibration of the analysis system was com
pleted on January 9, 2009.
Each period of time was assigned a group number to analyze for differences in data between
groups as shown in Table 13.
30
Table 13. Events Identified as Possible Data Shift Correlates

Group Description
Date
Number
9/30/08 to Highway FET cycle was run in triplicate, the exhaust gasses for analysis
1
11/18/08 from the runs were collected sequentially into Bag #1, Bag #2 and Bag#3
The fuel economy lab began a check of the valves and bags to determine if
11/19/08 2
there were mechanical differences
11/20/08 to
3 The air conditioning SC03 cycle data was significantly higher than expected
11/25/08
01/13/09 to New valves installed and entire system recalibrated. Selected tires were re
4
01/23/09 run on all cycles except Cold City FTP
10/21/08 to Cold City FTP cycle – initial tire types
5
11/06/08
12/02/08 to Cold City FTP cycle – additional tire types
6
12/17/08
3.1.2 Highway FET Triplicate Analysis:

The vehicle exhaust gas is routed to a number of bags for collection. For the tires in Group 1, the
runs for the triplicate analysis of the Highway FET cycle were compared. There was a significant
difference between the mileage for collection bags, with the mileage for bag #2 being approxi
mately 0.7 mpg (2%) higher than the values for bags #1 and #3. This is evident in the data shown
in Figure 14. The fuel economy lab investigated the mechanical functioning of the system on
November 19, 2008, and was unable to identify any equipment that was not functioning within
specification. The decision was made to complete the final group of tires without making any
changes to the equipment or procedures. Table 14 shows the analysis of variance (ANOVA) for
the Highway FET cycle for all tires in groups 1 to 3. On November 25, several runs were made
that showed the difference between values correlated to the physical collection bag used in the
analysis. Six tire types were re-run on the dynamometer in January 2009 after physical replace
ment of the valves in the system and provided data with equal values from all bags. Data for the
Highway FET was therefore analyzed by bag. Unfortunately, this data offset precludes the in
tended use of the Highway FET values to study the precision of the test method.
31
Exhaust Gas & Ambient Air

Sample Collection Bags
Figure 13. Vehicle Fuel Economy Dynamometer Exhaust Collection Bags and Control
System
mpg
39
38
37
36
35
34
39720 39730 39740 39750 39760 39770
Microsoft Date Code

Figure 14. Highway FET Schedule Fuel Economy Versus Bag Collection Number
32
Table 14. Analysis of Variance for Highway FET Fuel Economy by Tire Type and
Collection Bag Number
Dependent Variable: mpg
Source DF Sum of Squares Mean Square F Value Pr > F
Model 23 100679.7092 4377.3787 59474.9 <.0001
Error 53 3.9008 0.0736
Uncorrected Total 76 100683.6100
R-Square Coeff Var Root MSE mpg Mean
0.925213 0.745556 0.271294 36.38816
Source DF Type III SS Mean Square F Value Pr > F
type 20 40.59476016 2.02973801 27.58 <.0001
bag 2 7.64826645 3.82413323 51.96 <.0001
Duncan's Multiple Range Test for mpg
Alpha 0.05
Error Degrees of Freedom 53
Error Mean Square 0.0736
Harmonic Mean of Cell Sizes 25.32468
Means With the Same Letter

Are Not Significantly Different.
Duncan Grouping Mean N bag
A 36.81923 26 2
B 36.24400 25 3
C 36.08400 25 1
33
3.1.3 Air Conditioning SC03 – 11/20/08 to 11/25/08
Tire Type M14 (SRTT) was run as a control tire periodically throughout the test procedure. The
data on November 24, 2008, for the SC03 cycle was significantly higher than that seen previ
ously, as shown in Table 15. The same magnitude of difference was seen in a repeat of the G12
tire at 158 kPa. After the replacement of the valves and recalibration of the system was com
pleted on January 9, 2009, the data returned to the previous level. Figure 15 shows the data for
the SC03 cycle by group. It is apparent that the data from Group 3 does not follow the trend seen
in Group 1 for mpg by rolling resistance of the tire. After repair and recalibration of the system,
the trend for Group 4 returns to nearly that seen in Group 1. No apparent reason for the data shift
was seen, therefore SC03 data from Groups 2 and 3 were removed from the analysis and tire
types G11, P5 and R4 were repeated in Group 4.
Table 15. Air Conditioning SC03 Schedule, mpg for SRTT Tire by Date
G12 Type at 158 kPa, M14 Type at 210 kPa,
Date mpg mpg
09/30/08 19.3
10/15/08 19.3
10/17/08 19.4
11/10/08 20.0
11/2408 21.9
11/25/08 21.9
01/14/09 19.8
34
mpg
22.5
22.0
21.5
21.0
20.5
20.0
19.5
19.0
18.5
18.0
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 15. Air Conditioning SC03 Fuel Economy Versus Tire Rolling Resistance by
Analysis Group
3.1.4 Analysis by Date for Possible Drift in Data over Time

Tire types M14 (the SRTT) was repeated systematically during the testing as shown in Table 12.
The data for Group 1, or Groups 5 and 6 for the Cold City FTP cycle was analyzed by date of
testing to see if there was a drift in the data over time. Such a shift could result from systematic
changes in the vehicle or dynamometer. Table 16 summarizes the results of the analyses. Al
though no individual change in fuel economy is statistically significant, the fact that all values
seemed to increase slightly over the course of the experiment indicates that there was a slight
drift toward higher fuel economy over time. For the 50 days of testing in this analysis this would
result in a total increase of approximately 0.5 mpg. This means that pair-wise comparisons be
tween tests performed at different times would be suspect. For this analysis, the only pair-wise
comparisons that were studied were between the same tires at different inflation pressures. These
tests were normally performed on successive days, and the greatest difference between tests was
9 days. For the overall effect of tire rolling resistance, the rolling resistance of the tires studied
did not vary systematically over the test period, as seen in Figure 16, so this drift will constitute a
part of the error term in the analysis.
35
Table 16. Change in Fuel Economy Over Total Time of Testing

Test Probability > |t| Coefficient, mpg/day
Highway FET (Bag #1) 0.230 0.0221

Highway FET (Bag #2) 0.246 0.0035
Highway FET (Bag #3) 0.779 0.0045
City FTP 0.655 0.0033
High Speed US06 0.098 0.0095
Air Conditioning SC03 0.236 0.0182
Cold City FTP 0.094 0.0005
Average 0.0088
ISO 28580 RR, lbs

16
15
14
13
12
11
10
9
39720 39740 39760 39780 39800 39820 39840
Day, Microsoft Date Code
Figure 16. Rolling Resistance of Tires Tested Versus Day of Testing
Table 17 summarizes the data that was excluded from the individual analyses of effects on fuel
economy based on the preliminary data quality analysis.
36
Table 17. Data Excluded from Fuel Economy Analyses
Test Data Excluded
Highway FET (Bag #1) Group 4
Highway FET (Bag #2) None
Highway FET (Bag #3) Group 4
City FTP None
High Speed US06 None
Air Conditioning SC03 Groups 2 and 3
Cold City FTP None
Effect of Inflation Pressure None
Although a number of data quality issues were identified during the fuel economy testing, it is
important to stress that these problems were accounted for prior to ANOVA analysis. Therefore,
the fuel economy results presented in the report are believed to be accurate, as evidenced by
there relative agreement with results of similar studies contained in the literature.
3.1.5 Effect of Tire Rolling Resistance on Fuel Economy

Table 18 shows the results of the analysis of variance for the various fuel economy tests studied.
Figure 17 through Figure 23 illustrate the trends and 95 percent confidence limits of each test for
miles per gallon, calculated for four tires of the specified rolling resistance mounted on the vehi-
cle, versus the rolling resistance force values for the tires studied. Although the R2 values are
poor, the tire rolling resistance term is significant and the F Values indicate that the overall trend
toward lower fuel economy with increasing tire rolling resistance is statistically significant. Val-
ues of Probability |t| less than 0.05 indicate that the variable (lbf) has a significant effect on roll-
ing resistance. Figure 24 shows the fuel economy as a percentage of the mean for each test ver-
sus the rolling resistance as a percentage of the mean rolling resistance. The scatter in the data is
evident, but the overall trends are clear and the percentage decreases in fuel economy as tire roll-
ing resistance increases show very similar trends. Previous studies have shown the effects of roll-
ing resistance as percentage change in mileage for the vehicle (%mpg) that results from some
percentage change in rolling resistance of the tires (%Rolling Resistance). Table 19 shows the
regression results for this measure. The increase in mpg for a 10 percent decrease in rolling resis-
tance is approximately 1.1 percent, ranging from a low of 0.8 percent for the air conditioning
SC03 cycle to a high of 1.3 percent for the High-Speed US06 cycle. These results agree with the
calculated values of a 0.7 percent to 2.0 percent change in fuel economy for a 10 percent change
in rolling resistance that are shown in the Transportation Research Board, Special Report
286.[27]
37
Table 18. ANOVA Results for Effect of Tire Rolling Resistance on Fuel Economy
Probability Coefficient, mpg / lbf. Probability
Test F Value R2 Value
>F Rolling Resistance > |t|
Highway FET (Bag #1) 46.2 0.0001 0.687 -0.306 0.0001
Highway FET (Bag #2) 71.5 0.0001 0.695 -0.315 0.0001
Highway FET (Bag #3) 75.5 0.0001 0.770 -0.376 0.0001
Average Highway FET -0.332
City FTP 48.5 0.0001 0.593 -0.176 0.0001
High Speed US06 48.6 0.0001 0.611 -0.233 0.0001
Air Conditioning SC03 16.0 0.0005 0.381 -0.131 0.0005
Cold City FTP 45.7 0.0001 0.729 -0.168 0.0001
Table 19. Percentage Change in Fuel Economy Versus Percentage Change in Tire Rolling
Resistance
Coefficient, Variability,
Test
% mpg / % Rolling Resistance CV
Highway FET (Bag #1) -0.105 1.13
Average Highway FET -0.113 1.12
City FTP -0.102 1.40
High Speed US06 -0.132 1.77
Air Conditioning SC03 -0.083 1.78
Cold City FTP -0.112 1.07
38
mpg
38
37
36
35
34
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 17. Highway FET (Bag #1) Mileage Versus Tire Rolling Resistance
mpg
39
38
37
36
35
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
39
mpg
38
37
36
35
34
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
mpg
22.2
22.0
21.8
21.6
21.4
21.2
21.0
20.8
20.6
20.4
20.2
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 20. City FTP Mileage Versus Tire Rolling Resistance
40
mpg
23
22
21
20
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 21. High Speed US06 Mileage Versus Tire Rolling Resistance
mpg
20.4
20.2
20.0
19.8
19.6
19.4
19.2
19.0
18.8
18.6
18.4
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 22. Air Conditioning SC03 Mileage Versus Tire Rolling Resistance
41
mpg
19.3
19.2
19.1
19.0
18.9
18.8
18.7
18.6
18.5
18.4
18.3
18.2
18.1
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 23. Cold City FTP Mileage Versus Tire Rolling Resistance
42
mpg, %
105
104
103
102
101
100
99
98
97
96
95
94
70 80 90 100 110 120 130
Rolling Resistance, %
1 = Highway FET (Bag #1) 4 = City FTP 7 = Cold City FTP
2 = Highway FET (Bag #2) 5 = High Speed US06
3 = Highway FET (Bag #3) 6 = Air Conditioning SC03
Figure 24. Percentage Change in Fuel Economy Versus Percentage Change in
Tire Rolling Resistance
3.1.6 Effect of Reduced Inflation Pressure on Fuel Economy

Underinflated tires have been shown to be a prevalent issue for passenger vehicle safety. In
2001, NHTSA released the results of the Tire Pressure Special Study, showing that 28 percent of
passenger cars had at least one tire underinflated by 8 psi or more.[28] Recently NHTSA pub
lished the results of a sample of vehicles surveyed with and without tire pressure monitoring sys
tems (TPMS).[29] Although the number of vehicles with underinflated tires was less with
TPMS, there were still approximately 20 percent of vehicles equipped with TPMS systems that
had at least one tire underinflated by 25 percent or more. 8 While the primary safety issue for un
der-inflation of tires is reduced vehicle control and possible tire failure due to cumulative dam
age, studies have shown that underinflation increases the rolling resistance of a tire, thus increas
ing vehicle fuel consumption.
Using available literature (1991), Wicks and Sheets assumed that the rolling resistance of a tire
increased inversely with the “under pressure” ratio (actual tire pressure divided by the recom
mended tire pressure). Using a theoretical drop in pressure from 35 psi (241 kPa) to 25 psi (172
8
For those vehicles equipped with direct TPMS, there were approximately 1% percent with at least one tire underin
flated by 25% percent or more.
43
kPa), they estimated a 3.84 percent increase in vehicle fuel consumption for a 10 psi reduction in
tire inflation pressure.[30] The U.S. Department of Energy predicts an increase in fuel consump
tion of 3 percent for each 10 psi reduction in tire pressure.[31] Clark et al. show increases in roll
ing resistance of 1 percent to 3 percent for each psi reduction in inflation pressure.[32] Hall and
Moreland show increases in rolling resistance of 1 to 2 percent for a 1 psi reduction in inflation
pressure [33] and Continental Tire shows an increase in rolling resistance of 1.6-percent per psi
reduction in rolling resistance.[34] Using the NAS predictions of a 1- to 2-percent decrease in
fuel consumption for a 10 percent increase in rolling resistance, the predicted values in
%mpg/psi is shown in Table 20.
In 1981, the U.S. Environmental Protection Agency published an SAE paper on the effects of
reduced inflation pressure on fuel mileage measured during on-road vehicle testing.[35] The test
ing used two identical 1979 Chevrolet Nova passenger cars with the OEM radial tires. The two
vehicles were simultaneously driven on a closed test track through repetitive cycles of the then
current EPA Urban and Highway driving schedules. The vehicles had a placard tire pressure of
24 psi (165 kPa), and all four tires on the vehicle were either adjusted up or down by 4 psi (28
kPa) from the placard. The 8-psi (55-kPa) difference in tire pressure generated an average com
posite 9 fuel consumption to tire pressure change ratio of 0.33%/psi.
Table 20. Predicted Change in Fuel Economy for 1 psi Change in Tire Inflation Pressure
Study Predicted Reduction, Calculation Notes:
%mpg/psi
Wicks and Sheets 0.38 3.8%mpg/10psi
U.S. Department of Energy 0.3 3%pmg/10psi

1/%mpg/10%RR/1%RR/psi to
Clark et al. 0.1 to 0.6
2%/mpg/10%RR/3%RR/psi
1/%mpg/10%RR/1%RR/psi to
Hall and Moreland 0.1 to 0.4
2%/mpg/10%RR/2%RR/psi
1/%mpg/10%RR/1.6%RR/psi to
Continental Tire 0.16 to 0.32
2%/mpg/10%RR/1.6%RR/psi
Measured in on-vehicle tests,
U.S. EPA 0.33
published 1981
Average 0.308
Referencing Schuring’s work in 1980, LaClair states[36]:
“When a tire is under-inflated, its rolling resistance increases by a factor of, on average,
about (P/P0)-0.4, where P0 is the specified inflation pressure.”
Per this relationship, a (158/210)-0.4 or 12 percent increase in rolling resistance should be ob
served. Also referencing Schuring, NAS states[37]:
9
Composite fuel cycle was weighted for 55 percent urban cycles and 45 percent highway cycles.
44
“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.”
Per this relationship, a 10-percent increase (30-23 psi = 7 psi * 1.4%/psi = 10 %) in rolling resis
tance should be observed.
The effect of reduced inflation pressure was estimated from comparison of the dynamometer fuel
economy of the vehicle with the front tires inflated to the placard pressure of 210 kPa (30 psi), to
tests with the same front tires inflated to 158 kPa (23 psi). The lower pressure represents the 25
percent reduced pressure activation threshold of the tire pressure monitoring system (TPMS)
specified in FMVSS No. 138. Six tire models that spanned the range of rolling resistances were
chosen for the experiment. Unlike on-road vehicle operation, tests on the indoor fuel economy
dynamometer involve only the driven axle(s) of the vehicle, which for the Chevy Malibu test ve
hicle was the front axle. Since the EPA fuel economy test is completed using only the driven
axle, it is assumed that the effects of the drag of the non-driven axle (rear axle) on vehicle fuel
economy are accounted for in the EPA’s complex equations. Therefore, it is also assumed that
the increased rolling resistance of the front tires due to underinflation will scale up through these
equations to the effects of four underinflated tires on vehicle fuel economy. However, no on-road
testing was completed to confirm this.
Statistical pair-wise comparisons of the same tires tested at the two different inflation pressures
did not show significant differences in fuel economy. However, there was a trend for tires at the
lower inflation pressure to generate lower fuel economy in all tests as shown in Figure 26 to
Figure 32, which illustrates the data by tire type, where the “L” suffix (e.g., M14L) indicates the
158-kPa inflation pressure. Table 21 shows the results of the ANOVA analysis for the tests. All
but one of the tests showed a decrease of 0.3 to 0.5 miles per gallon for all fuel economy cycles
for the 25-percent decrease in tire pressure. The High Speed US06 test showed no significant
change in fuel economy.
Using the relationship between rolling resistance and fuel economy for the vehicle tested by
NHTSA as shown in Table 19, a 0.11-percent reduction in fuel economy is predicted for each 1
percent increase in rolling resistance. Based on LaClair’s estimate of rolling resistance increases
of 12 percent for the reduced pressure, a 1.3-percent reduction in overall fuel economy for all
cycles is predicted. Therefore, the actual results in Table 21 of an approximate 1.17-percent re
duction in overall fuel economy for all cycles is fairly close to the 1.3-percent prediction. The
predicted value using the NAS estimate of a 1.1-percent decrease in fuel economy is also very
close to the measured value of 1.17 percent.
However, the measured fuel economy of the vehicle at reduced inflation pressure was signifi
cantly less than the average predicted value shown in Table 20. When weighted in the same
manner as used by the EPA, the results for the Urban and Highway cycles in Table 19 predict
fuel consumption to tire pressure change ratio of 0.11 percent/psi, or about a third of that seen by
the EPA on the track in 1979. The following explanations, or combinations of the following, are
thought to be possible:
45
1. The effects of underinflation on the tire contact patch that would occur with a tire on a
relatively flat road may not approximated by the Hertzian-like contact of the tire on the
curved 48-inch diameter steel rollers of the fuel economy dynamometer (Figure 25).
2. The increased heat build-up of the low inflation tires on the dynamometer rollers, which
could raise the inflation pressure significantly over the short duration of the test and
lower the differential in rolling resistance between the under-inflated and properly in
flated tires.
3. Assuming a nominal load on the tires of 80 percent of maximum sidewall, the predicted
10 percent increase in rolling resistance due to 25 percent underinflation for the highest
rolling resistance tire model in the study (type B12) is 15.22 lbf x 0.1 x 2 tires = +3.0 lbf
for two tires on the front axle. For the lowest, the OE type G12 tires, it is 9.47 x 0.1 x 2
tires = +1.9 lbf. In horsepower terms, at the 60 mph speed of the highway test cycle the
increased rolling resistance forces are roughly 0.3-0.5 hp on a vehicle with an engine
rated at 211 hp at 5800 rpm, or only 0.14-0.24 percent of maximum horsepower. The ve
hicle’s modern fuel, ignition, and powertrain management software, which may include
adaptive spark timing, adaptive exhaust catalyst, mass-airflow sensors, and other tech
nologies to lessen emissions and optimize fuel economy, may mitigate some of the ef
fects of the additional rolling resistance of the two underinflated tires at the front axle.
This is particularly significant when comparing these results to earlier results using car
bureted vehicles. For instance, the carbureted 1979 Chevrolet Nova vehicles used in the
EPA testing were rated at 115 maximum horsepower at 3800 rpm.
4. The dynamometer loading is dependent on the road load coast-down coefficients of the
vehicle with the OE tires at placard cold inflation pressure. This in turn affects the emis
sions results and therefore the fuel economy of the vehicle. New road load coast-down
coefficients may have to be determined for vehicle tests at the lower TPMS activation
pressure.
Figure 25. Tire to Dynamometer Roller Contact / 2008 Chevrolet Impala LS Engine
46
mpg
40
38
36
34
32
30
B B B B G G G G M M M M
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
Figure 26. Highway FET (Bag #1) Fuel Economy by Tire Type and Inflation Pressure
mpg
40
38
36
34
32
30
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
47
mpg
40
38
36
34
32
30
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
mpg
25
23
21
19
17
15
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
Figure 29. City FTP Fuel Economy by Tire Type and Inflation Pressure
48
mpg
25
23
21
19
17
15
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
Figure 30. High Speed US06 Fuel Economy by Tire Type and Inflation Pressure
mpg
20
18
16
14
12
10
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
Figure 31. Air Conditioning SC03 Fuel Economy by Tire Type and Inflation Pressure
49
mpg
20
18
16
14
12
10
1 1 1 1 1 1 8 8 1 1 1 1
1 1 3 3 2 2 L 3 3 4 4
L L L L L
type
Figure 32. Cold City FTP Fuel Economy by Tire Type and Inflation Pressure
Table 21. ANOVA Results for Effect of Tire Inflation Pressure Reduction on Fuel Economy
Coefficient Probability
Test
mpg / 52 kPa %mpg / kPa %mpg / psi > |t|
Highway FET (Bag #1) -0.553 -0.030 -0.203 0.0119
Highway FET (Bag #2) -0.173 -0.009 -0.062 0.0164
Highway FET (Bag #3) -0.206 -0.011 -0.075 0.0092
Average Highway FET -0.311 -0.0167 -0.113 0.0125
City FTP -0.166 -0.015 -0.103 0.0008
High Speed US06 -0.018 -0.001 -0.011 0.8200
Air Conditioning SC03 -0.497 -0.049 -0.337 0.0096
Cold City FTP -0.305 -0.031 -0.216 0.0088
3.1.7 Fuel Economy Testing Summary

The repeatability of the fuel economy tests were found to range from 1 to 2 percent. Using a
brand new vehicle, with mileage break-in to the prescribed 2,000 miles, a significant upward
drift in the average mpg between 2,000 and 6,000 miles was noted. Offsets between the intended
triplicate Highway FET results precluded a more precise assessment of the accuracy of the test.
In spite of these limitations, there was a significant effect of tire rolling resistance on fuel econ
omy for all tests over the 9.47 to 15.60 pound range of the tires studied. For all tests, a 10 percent
decrease in rolling resistance resulted in slightly more than a 1 percent increase in fuel economy
for the vehicle. Reducing the inflation pressure by 25 percent resulted in a small but statistically
significant decrease in the fuel economy of the vehicle by approximately 0.3 to 0.5 miles per gal
50
lon for all tests except the high-speed, high-acceleration US06 cycle. Figure 33 illustrates the
trends for the Highway FET (Bag #2) test. As before, the suffix of “L” for the tire type indicates
the low-pressure condition. The trend to lower mileage with increased rolling resistance is clear.
In general, the tires at lower inflation pressure have lower gas mileage in line with their calculat
ed 10 rolling resistance at the reduced pressure.
mpg
39
38
37
36
35
9 10 11 12 13 14 15 16
ISO 28580 RR, lbs
Figure 33. Highway FET (Bag #2) Fuel Economy Versus Tire Rolling Resistance by Tire
Type and Inflation Pressure
3.2 Correlation of Tangent δ at 60°C to Tire Rolling Resistance

The tread compounds of 16 tire models were tested for dynamic mechanical properties. Litera
ture has suggested that the tangent δ at 60°C correlates to the tread’s contribution to the rolling
resistance of a tire. The Pearson R product moment correlation of the measures for the 16 tires is
shown in Table 22. The measured rolling resistance has a significant positive correlation of 0.72
to the tangent δ measured in both shear and tension. The measures in shear and tension are very
highly correlated. The data for rolling resistance versus tangent δ measured in tension by tire
type is shown in Figure 34. It appears that the data falls into two groups of tangent δ, which
strongly influences the overall correlation. However, the low tangent δ and low rolling resistance
group consists of two tire brands. Interestingly, these three tire types shown in the lower left cor
ner of the graph have all been confirmed by their manufacturers as being original equipment tires
models. Table 23 shows the correlation of rolling resistance to some compound properties and
10
Calculated from the coefficients of the multi-point rolling resistance tests of those tire types.
51
how they differ for two brands of tires. Both the Bridgestone and Goodyear tires individually
show a stronger correlation of rolling resistance and tangent δ at 60°C than that shown for all
tires grouped together. The differing correlations to compound properties may be related to dif
fering compound strategies of the companies to tailor compounds for individual tire perform
ance. The Bridgestone tires seem to be sensitive to increases in the total filler level, while the
Goodyear tires seem to be more sensitive to the type of filler (percent silica) and possibly to
other compounding ingredients (volatile content). A broad range of tire types were measured for
compound properties and rolling resistance and will be reported separately.
Table 22. Correlation of Rolling Resistance to Tangent δ at 60°C

Correlation to Rolling Tangent δ at 60°C Tangent δ at 60°C
Resistance (Tension) (Shear)
0.7216 0.7311
Probability > |r| 0.0016 <0.0001
Rolling Resistance
16
15
14
13
12
11
10
0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22
Tangent delta @60C in Tension

Figure 34. ISO 28580 Rolling Resistance (lbs)Versus Tangent δ at 60°C by Tire Type
OE Tires
OE Run-Flat Tire
52
Table 23. Correlation of Properties to Rolling Resistance

Parameter Correlation to Rolling Resistance
All Tires Bridgestone Tires Only Goodyear Tires Only
Tangent δ at 60°C (tension) 0.722 0.901 0.932
Percent Polymer in Compound -0.816 -0.920 -0.598
Percent Volatiles in Compound 0.645 0.828 0.912
Total Filler Level of Compound 0.814 0.907 0.341
Percent of Filler That is Silica -0.109 -0.068 -0.945
3.3 Effect of Tire Rolling Resistance on Safety

Sixteen tire models representing a range of rolling resistance and of other characteristics were
tested for both dry and wet traction by NHTSA. The testing was done in conjunction with the
standard ASTM E501 Standard Reference Test Tire used for UTQGS testing. The FMVSS
575.104 UTQGS Traction grade is based upon the wet slide traction coefficient, which is ad-
justed with results from the ASTM E501 tire to correct for variations in the test surface, envi-
ronmental conditions, etc. Since the standard does not have a procedure for adjusting wet peak,
or dry peak and slide traction coefficients, it was decided to report raw traction data as “Traction
Number” (coefficient of friction x 102) and as a “Ratio to the Course Monitoring Tire” (ASTM
E501). The ratios were calculated using the only results from the E501 course monitoring tires
used for that tire model’s traction testing sequence. In this manner, the ratios represent data that
has been “corrected” for variations in the test surface, environmental conditions, etc. The wet
slide data is reported in this same format, as well as in the form of the UTQGS adjusted traction
coefficient. Since these tires were previously tested for rolling resistance, and did not undergo
the full break-in period, the results for the adjusted traction coefficients are unofficial. The raw
peak and slide traction data for dry and wet, asphalt and concrete testing, as well as adjusted
UTQGS wet slide coefficients are contained in Appendix 4 through Appendix 10.
3.3.1 Dry Traction Data

Table 24 shows the average dry traction results for each tire type on asphalt and concrete. Table
25 shows the Pearson Product Moment Correlation of the values for dry traction to the tire roll-
ing resistance. The Pearson value indicates the strength and direction of the correlation with val-
ues ranging from -1 for complete inverse correlation, to +1 for complete direct correlation, with
values near zero indicating no correlation between the measures. It is evident that there is very
little correlation between the traction and rolling resistance for these tires. For a value to be sta-
tistically significant the probability > |r| would have to be less than 0.050, and no value ap-
proaches that number. Figure 35 and Figure 36 display clearly that there is no indication that a
tire with improved rolling resistance will necessarily have lower dry traction performance in this
test.
53
Table 24. Dry Traction Results, Traction Number and Ratio to E501 Reference Tire
Tire ISO 28580 Roll Traction
Type ing Resistance, Asphalt Concrete
lbf Peak Value Sliding Value Peak Value Sliding Value
Traction Ratio Traction Ratio Traction Ratio Traction Ratio
Number E501 Number E501 Number E501 Number E501
G8 9.83 94.41 95 65.95 110 93.25 88 75.31 95
G11 10.02 97.45 99 64.66 93 104.07 96 75.95 93
B11 10.13 94.77 96 60.73 98 101.12 93 74.43 91
G9 11.27 98.25 98 74.16 109 102.20 95 78.82 97
M14 11.96 99.53 101 66.67 104 105.50 97 81.70 100
M13 12.07 100.12 101 53.75 82 105.62 97 69.66 85
G10 12.09 98.53 96 74.00 101 102.07 94 78.39 97
B10 12.11 93.83 94 77.65 127 96.45 91 86.63 107
D10 13.56 94.60 95 62.10 101 102.71 96 74.77 94
B14 13.90 101.50 102 75.76 125 107.58 100 85.02 106
U3 13.91 91.75 94 67.23 108 100.22 93 79.71 103
B15 13.99 90.64 92 66.99 107 91.93 86 75.42 97
P5 14.02 95.61 95 56.97 96 94.63 90 71.52 92
R4 14.98 104.19 106 71.13 112 107.86 103 84.38 104
B13 15.01 94.87 94 57.63 96 91.93 88 76.42 98
B12 15.22 103.90 106 56.33 89 108.18 102 71.95 88
E501 - 99.23 100 63.48 100 107.15 100 80.32 100
Table 25. Pearson Product Moment Correlation of Dry Traction to Rolling Resistance
Pearson Product Moment Correlation
Asphalt, Dry Traction Concrete, Dry Traction
Correlation to ISO
Peak Value Sliding Value Peak Value Sliding Value
28580 Rolling Re
sistance Traction Ratio Traction Ratio Traction Ratio Traction Ratio
0.209 0.200 -0.158 0.045 0.056 0.209 0.069 0.217
Probability > |r| 0.2518 0.2730 0.3886 0.8073 0.7602 0.2507 0.7059 0.2336
54
Figure 35. Dry Traction Numbers Versus ISO 28580 Rolling Resistance
Concrete - Peak Number

120
Asphalt - Peak Number
Concrete - Slide Number
110
Asphalt - Slide Number
2
Traction Number
R = 0.0033
100 2
R = 0.044
90
80 2
R = 0.0058
70
2
R = 0.0241
60
50
9 10 11 12 13 14 15 16
ISO 28580 Rolling Resistance (lbs)
55
Figure 36. Dry Traction Ratios to E501 Course Monitoring Tire Versus Rolling Resistance
135
Asphalt - Slide Ratio
Concrete - Slide Ratio
125 Asphalt - Peak Ratio
Ratio to ASTM E501 Value
Concrete - Peak Ratio
115
105 2
R = 0.0003
2
R = 0.047
2
R = 0.0445
95 2
R = 0.0469
85
75
9 10 11 12 13 14 15 16
3.3.2 Wet Traction Data

Table 26 shows the average wet traction results for each tire type on asphalt and concrete. Table
27 shows the Pearson Product Moment Correlation of the values for wet traction to the tire roll
ing resistance. The Pearson value indicates the strength and direction of the correlation with val
ues ranging from -1 for complete inverse correlation, to +1 for complete direct correlation, with
values near zero indicating no correlation between the measures. For a value to be statistically
significant the probability > |r| should be less than 0.050. The sliding values all have a strong and
significant relationship between better rolling resistance and poorer wet traction. The peak values
display the same tendency but the relationship is much weaker.
Figure 37 and Figure 38 display these trends graphically for the traction numbers and the ratio to
the E501 monitoring tire respectively. From these data, it appears that the tires with lower rolling
resistance values will have poorer wet traction performance in the sliding region. This will be
particularly significant to consumers without ABS systems on their vehicles, since the sliding
value will relate most closely to emergency stopping maneuvers. In contrast, the results for the
measured wet peak traction number in the same figures exhibit much less pronounced trends.
Hence, for newer vehicles with ABS or ESC systems, the tradeoff is expected to be much less
significant.
56
Table 26. Wet Traction Results, Traction Number and Ratio to E501 Reference Tire
Wet Traction
ISO 28580 Asphalt Concrete
Tire
Rolling Re Peak Value Sliding Value Peak Value Sliding Value
Type
sistance, lbs Traction Ratio Traction Ratio Traction Ratio Traction Ratio
G8 9.83 87.6 101 48.9 93 58.9 103 35.1 100
G11 10.02 82.9 96 49.9 95 63.4 111 36.6 104
B11 10.13 87.2 102 46.4 90 63.0 110 36.4 99
G9 11.27 82.2 101 54.7 102 58.6 102 36.4 102
M14 11.96 94.8 104 58.8 109 66.2 116 39.6 109
M13 12.07 93.8 103 50.9 97 73.4 132 40.1 111
G10 12.09 83.5 105 55.1 101 56.3 106 36.7 103
B10 12.11 80.0 95 49.5 92 48.6 90 37.4 104
D10 13.56 89.3 106 54.5 100 68.2 122 39.5 109
B14 13.90 94.4 108 58.9 111 76.2 128 42.2 115
U3 13.91 87.5 100 53.7 100 64.9 109 40.2 109
B15 13.99 79.3 94 52.4 97 54.1 101 35.4 98
P5 14.02 84.1 99 54.3 105 70.2 124 41.0 112
R4 14.98 86.9 103 60.5 111 64.5 115 39.1 107
B13 15.01 92.3 105 57.7 108 71.1 120 41.0 111
B12 15.22 96.0 118 59.1 110 80.1 140 42.3 119
E501 - 85.8 100 53.3 100 56.4 100 36.1 100
Table 27. Pearson Product Moment Correlation of Wet Traction to Rolling Resistance
Asphalt, Wet Traction Concrete, Wet Traction
Correlation to ISO 28580
Rolling Resistance
Traction Ratio Traction Ratio Traction Ratio Traction Ratio
0.299 0.391 0.739 0.725 0.465 0.473 0.700 0.628
Probability > |r| 0.0965 0.0270 <0.001 <0.001 0.007 0.006 <0.001 0.001
57
Figure 37. Wet Traction Numbers Versus ISO 28580 Rolling Resistance
110
Asphalt - Peak Number
Concrete - Peak Number
100
Asphalt - Slide Number
2
Concrete - Slide Number R = 0.0875
90
Traction Number
80
2
R = 0.2116
70
2
R = 0.5511
60
50
2
R = 0.5035
40
30
9 10 11 12 13 14 15 16
58
Figure 38. Wet Traction Ratios to E501 Course Monitoring Tire Versus Rolling Resistance
145
Concrete - Peak Ratio
Concrete - Slide Ratio
135 Asphalt - Slide Ratio
Asphalt - Peak Ratio
Ratio to ASTM E501 Value
125 2
R = 0.2273
115 2
R = 0.4133
2
R = 0.524
2
105 R = 0.1657
95
85
9 10 11 12 13 14 15 16
3.3.3 UTQGS Traction Grade

As stated earlier, the FMVSS 575.104 UTQGS Traction grade is based upon the adjusted traction
coefficient, which is calculated from the candidate tire’s wet slide coefficient of friction and the
standard reference tire’s wet slide coefficient of friction using the following procedure:
Average the 10 measurements taken on the asphalt surface to find the candidate tire trac
tion coefficient for the asphalt surface. Average the 10 measurements taken on the con
crete surface to find the candidate tire traction coefficient for the concrete surface.
(ix) Compute a candidate tire’s adjusted traction coefficient for asphalt (μa) by the
following formula:
μa = Measured candidate tire coefficient for asphalt + 0.50 - Measured

standard tire coefficient for asphalt.
(x) Compute a candidate tire’s adjusted traction coefficient for concrete (μc) by
the following formula:
μc = Measured candidate tire coefficient for concrete +0.35 μMeasured

standard tire coefficient for concrete.
59
The results for the UTQGS Adjusted Traction Coefficient (unofficial) on asphalt verses the roll
ing resistance of the tire as measured by ISO 28580 are presented in Figure 39. The limits for the
three grades within the span of the test tires are indicated on the figure. Similar to the raw wet
traction trailer data, the adjusted traction coefficient on asphalt was lower in lower rolling resis
tance tires.
Figure 39. UTQG Adjusted Traction Coefficient for Asphalt Versus ISO 28580 Rolling
Resistance
AA Traction
Grade
0.57
Adjusted Traction Coefficient (Mua)
A Traction
0.52
Grade
R2 = 0.5175
0.47
B Traction
Grade
0.42
0.37
9 10 11 12 13 14 15 16
The results for the UTQGS Adjusted Traction Coefficient (unofficial) on concrete verses the
rolling resistance of the tire as measured by ISO 28580 are presented in Figure 40. The limits for
the three grades within the span of the test tires are indicated on the figure. Again, similar to the
raw wet traction trailer data, the adjusted traction coefficient on concrete was lower in lower roll
ing resistance tires.
60
Figure 40. UTQG Adjusted Traction Coefficient for Concrete Versus ISO 28580 Rolling
Resistance
0.43
Traction
Grade
0.41
AA
R2 = 0.405
Adjusted Traction Coefficient (Muc)
0.39
Traction
Grade
0.37
A
0.35
0.33
Traction
Grade
0.31
B
0.29
0.27
0.25
9 10 11 12 13 14 15 16
3.3.4 Correlation of Tangent δ at 0°C to Wet Traction Properties

Literature has shown a general trend towards increased wet traction properties of tires as the tan
gent δ at 0°C of the tread compound increases. Table 28 shows the Pearson correlation coeffi
cient between the measured slide numbers for wet traction and the tangent δ at 0°C of the tread
compound measured in tension. There is a strong positive correlation of the tangent δ at 0°C to
wet traction of the tires, particularly for the sliding values. The data for the slide number on con
crete versus tangent δ at 0°C is shown in Figure 41. This relationship appears to be more gener
ally applicable than that seen for rolling resistance to tangent δ at 60°C.
61
Table 28. Pearson R Product Moment Correlation of Wet Traction to
Tangent δ at 0°C of the Tread Compound

Correlation Asphalt, Wet Traction Concrete, Wet Traction
to Tangent δ
at 0°C (Ten
Traction Ratio Traction Ratio Traction Ratio Traction
sion) Ratio E501
Number E501 Number E501 Number E501 Number
0.6653 0.8119 0.8432 0.8635 0.7381 0.7536 0.8375 0.8556
Probability >
0.0049 0.0001 <0.0001 <0.0001 0.0011 0.0007 <0.0001 <0.0001
|r|
Slide Number / Concrete

43
42
41
40
39
38
37
36
35
0.16 0.18 0.20 0.22 0.24 0.26 0.28 0.30 0.32 0.34 0.36 0.38 0.40
Tangent delta @0C in Tension

Figure 41. Slide Traction Number on Wet Concrete Versus Tangent δ at 0°C Measured in
Tension
OE Tires
OE Run-Flat Tire
3.4 Effects of Tire Rolling Resistance on Treadwear Rate

As was seen previously, there was not a good correlation between the rolling resistance and the
UTQGS treadwear grade of the tires studied. Four tire models that were selected to represent the
range of rolling resistance of the models studied, along with the SRTT (tire type M14), were
tested according to the UTQGS testing protocol for treadwear. Although these tires were previ
ously tested for rolling resistance in a laboratory, the wear rates and projected mileages are ex
pected to be similar to those for new tires of the same model. Measurements were taken across
62
the tire at six locations in each groove (1 through 4). Data were analyzed by tire type, by groove,
by shoulder (groove 1&4) or tread center (groove 2&3). General observations are shown in
Table 29 for each tire type. The coefficients of variation for the wear rates are approximately 0.5
percent for all tire types indicating that comparisons between tire types at these conditions are
reliable. Models for the wear rate against course mileage produced R2 values of 0.94 to 0.97 for
linear models and 0.98 to 0.99 for quadratic models. For all tire types except B13 the quadratic
term was statistically significant, indicating that the wear rate tends to change (either increase or
decrease) as the tire wears.
Table 29. Analysis of Tire Wear Data

Tire Coefficient of Groove 1 to 4 Shoulder Versus Non-Linear
Type Variation Tread Center Behavior
B11 0.30% Groove 1 shows faster Shoulder wear rate faster Wear rate tends to
wear rate 11 than tread center increase
B13 0.44% - Similar wear rates No change in wear
rate
G8 0.51% Groove 4 shows slower Similar wear rates Wear rate tends to
12
wear rate increase
M13 0.54% - Tread center wear rate faster Wear rate tends to
than shoulder decrease
M14 0.43% - Tread center wear rate faster Wear rate tends to
than shoulder decrease
Table 30 shows the treadwear rates and projected mileage to 2/32nds tread depth for the tires. For
each model the wear rates for the shoulder and tread center were compared along with the pro
jected lifetime for each area. For tire type B11, the wear rate in the shoulder area was signifi
cantly faster than the wear rate in the tread center with a corresponding decrease in projected
mileage. For tire type M14 the wear rate in the tread center was significantly faster than in the
shoulder area with significantly shorter projected tread life in this area. Tire type M13 had faster
wear rates in the tread center but this was partially offset by a lesser groove depth in the shoulder
area in projecting tire lifetime.
Figure 42 shows the projected average tire mileage to wear out and the minimum projected mile
age, versus the rolling resistance for the tire. From these data, there is no relationship between
expected tire lifetime and rolling resistance. Since the tread depth may affect both rolling resis
tance and tire lifetime, the average wear rate and the fastest wear rate, either from the shoulder or
tread center area, was compared to the rolling resistance. It is evident from Figure 43 that there is
no clear relationship between wear rate and rolling resistance for these tires. In summary, there is
no evidence from this data that a tire with reduced rolling resistance will necessarily have re
duced tread life.
11
Data was influenced by high wear rate of tire #3146. The other B11 tires showed no anomalous behavior for indi
vidual grooves
12
All type G8 tires showed anomalous behavior for groove 4
63
Table 30. Wear Rates and Projected Mileage to 2/32nds Tread Depth
From UTQGS Treadwear Course
Reported Wear
Projected Life,
Projected Life,
tance, pounds
Rolling Resis-
Tread Center,
miles (Shoul-
Wear Rate in
Wear Rate in
miles (Tread
mil/1000mi
mil/1000mi
mil/1000mi
Tread Life,
Shoulder,
Projected
Tire Type
Center)
miles
Rate,
der)
B11 10.13 5.155 54,840 5.752 4.528 48,550 63,200
B13 15.01 6.463 52,020 6.374 6.276 51,790 54,540
G8 9.83 6.447 45,390 6.211 6.471 46,460 45,840
M13 12.07 5.448 41,310 4.795 5.768 45,150 40,500
M14 11.96 5.558 45,000 4.359 6.449 56,730 39,230
Figure 42. Projected Tire Mileage to Wearout (Average and Minimum) Versus ISO 28580
Rolling Resistance
UTQG Course Wear Versus Rolling Resistance
60,000
Average Miles
Minimum Miles
55,000
Projected Lifetime, miles
50,000
45,000
40,000
35,000
9 10 11 12 13 14 15 16
ISO 28580 Rolling Resistance, pounds
64
Figure 43. Average and Fastest Treadwear Rate Versus ISO 28580 Rolling Resistance
Wear Rates Versus Rolling Resistance

6.6
6.4
Wear Rate, mils / 1000 miles
6.2
5.8
Average
5.6 Fastest
5.4
5.2
5
9 10 11 12 13 14 15 16
3.4.1 Analysis of Wear Data From Indoor Treadwear Testing
Four tire models which were selected to represent the range of rolling resistance of the models
studied, along with the SRTT (Type M14) and the original equipment tires from the Chevrolet
Impala used in the fuel economy testing (Type G12), were tested using an experimental protocol
for indoor treadwear testing. While the computerized data acquisition provides 3-dimensional
measurements of the tire profile as shown in Appendix 3, representative measurements at 400
and 800 miles of wear were taken across the tire at six locations; the inside shoulder, the outside
shoulder, and at each of the 4 tread grooves. The average data for the models is shown in
Appendix 3. Tires were tested in duplicate at a severe condition, and for five of the models a
third tire was tested at a milder wear condition. Data was analyzed by tire type, by groove, by
shoulder or tread center and by wear severity.
General observations are shown in Table 29 for each tire type. Although all tires tended to show
somewhat faster wear in the shoulder region than at the tread center, tire type B11 showed wear
rates more than 3 times faster in the shoulder region than at the tread center. This tire type also
showed significantly faster wear rates in the shoulder area in the UTQGS course testing. As ex
pected, the wear rates at the severe condition were two to three times faster than at the mild con
65
dition. The loss per mile on this test was 3 to 20 times the wear rates that the same model tires
experienced on the UTQGS testing course.
Table 31. Indoor Treadwear Tire Wear Data

Average Wear Rate, Shoulder Versus Tread Center
Tire Type
mils/1,000 miles Wear Rate, ratio
Severe Condition Mild Condition Severe Condition Mild Condition
B11 103.1 28.8 3.3 3.2
B13 42.4 17.9 2.4 2.6
G8 101.5 37.1 2.3 1.6
G12 57.3 NA 3.1 NA
M13 28.8 12.7 1.2 0.9
M14 36.8 16.3 1.4 1.8
Table 30 shows the projected tire lifetime for severe and mild conditions calculated for both the
shoulder and tread center region of the tires. All tires had the shortest predicted lifetime in the
shoulder area. For the severe wear condition, this ranged from 1,250 miles for tire type B11 and
G8 to 5,500 miles for tire type M13. At the mild condition, the projected lifetime was two to
three times that of the severe condition. Figure 42 shows the projected tire lifetime versus rolling
resistance for the tire types and Figure 45 shows the wear rate versus tire rolling resistance. Both
indicate that there is a trend toward faster wear on this test for tires with lower rolling resistance.
The ANOVA analysis shown in Table 33 to Table 36 indicates that the relationship between
lower rolling resistance and lower expected tread life is statistically significant for this test. This
relationship is stronger for the projected lifetime based on the wear at the tread center, as evi-
denced by the R2 values of 0.75 to 0.8. Each decreased pound of original tire rolling resistance
correlates to approximately 1,000 miles of reduced wear at the mild condition and 2,000 miles at
the severe condition.
Table 32. Projected Mileage to 2/32nds Inch of Tread Depth

Projected Tread Life, miles
Tire
ISO 28580 Rolling Resistance, lbs Severe Condition Mild Condition
Type
Tread Center Shoulder Tread Center Shoulder
G12 9.47 5,156 2,035 NA NA
G8 9.83 4,137 1,264 9,200 5,327
B11 10.13 4,412 1,260 12,842 4,613
M14 11.96 7,732 4,937 17,476 8,738
M13 12.07 8,432 5,717 15,648 12,000
B13 15.01 10,902 4,440 20,628 9,114
66
25,000
Severe - Center
Severe - Shoulder
Mild - Center
20,000
Projected Lifetime, miles
Mild = Shoulder
15,000
10,000
5,000
0
9 10 11 12 13 14 15
Figure 44. Projected Tire Lifetime for Indoor Treadwear Test
Versus ISO 28580 Rolling Resistance
67
160
140 Severe - Center

Severe - Shoulder
Mild - Center
Wear Rate, mils / 1000 miles
120
Mild = Shoulder
100
80
60
40
20
0
9 10 11 12 13 14 15
Figure 45. Treadwear Rate for Indoor Treadwear Test
Versus ISO 28580 Rolling Resistance
68
Table 33. Projected Lifetime Versus Rolling Resistance – Mild Wear at Tread Center
Dependent Variable: Projected Lifetime at Tread Center
Wear Condition=Mild

Model 1 61222870.64 61222870.64 12.10 0.0401
Error 3 15183812.57 5061270.86
Corrected Total 4 76406683.21
R-Square Coeff Var Root MSE clife Mean
0.801276 14.84057 2249.727 15159.30
Rolling Resistance 1 61222870.64 61222870.64 12.10 0.0401
Parameter Estimate Standard Error t Value Pr > |t|
Intercept -9197.417585 7075.025908 -1.30 0.2845
Rolling Resistance 2045.064795 588.003644 3.48 0.0401
69
Table 34. Projected Lifetime Versus Rolling Resistance – Severe Wear at Tread Center
Dependent Variable: Projected Lifetime at Tread Center

Wear Condition=Severe
Model 1 58276986.51 58276986.51 28.46 0.0005
Error 9 18430645.35 2047849.48
R-Square Coeff Var Root MSE clife Mean
0.759729 20.53616 1431.031 6968.346
Intercept -6990.936132 2652.091689 -2.64 0.0271
70
Table 35. Projected Lifetime Versus Rolling Resistance – Mild Wear at Shoulder
Dependent Variable: Projected Lifetime at Shoulder

Wear Condition=Mild
Model 1 11993956.24 11993956.24 1.48 0.3114
Error 3 24391258.65 8130419.55
R-Square Coeff Var Root MSE slife Mean
0.329638 35.82655 2851.389 7958.872
Intercept -2821.735065 8967.155960 -0.31 0.7736
71
Table 36. Projected Lifetime Versus Rolling Resistance – Severe Wear at Shoulder

Dependent Variable: Projected Lifetime at Shoulder
Wear Condition=Severe
Model 1 15024648.61 15024648.61 5.28 0.0472
Error 9 25626840.14 2847426.68
R-Square Coeff Var Root MSE slife Mean
0.369597 48.39223 1687.432 3486.990
Intercept -3600.895727 3127.272986 -1.15 0.2792
Figure 46 shows the projected tread life for the tires versus the UTQGS tread grade. There is no
clear relationship between the UTQGS number and the projected tread life on this test.
72
14,000
12,000
Projected Miles to 2/32 Tread
Severe
Mild
10,000
8,000
6,000
4,000
2,000
0
250 350 450 550 650 750
UTQGS Treadwear Grade
Figure 46. Projected Tire Lifetime for Indoor Treadwear Test
Versus UTQGS Treadwear Grade
Change in Rolling Resistance Versus Tire Wear

The rolling resistance of each tire was measured using the ISO 28580 test procedure. All tires
were measured prior to beginning the testing. At the severe wear condition, the tires were also
measured after 400 miles and at the end of the 800-mile test. At the mild wear condition, the tires
were measured at the end of the 800-mile test. Tires were also weighed prior to the rolling resis
tance test and after 400 and 800 miles of wear. The data is shown in Appendix 3. Figure 47
shows the rolling resistance versus the weight of the tires. Some tires, such as type G12 and B11,
seem to show consistent decreases in rolling resistance with tread loss, and there seems to be a
tendency for lighter tires to have lower rolling resistance. However, many tires show no clear
relationship between tread loss and rolling resistance for this testing. Figure 48 shows the rolling
resistance as a percent of the original value for each tire versus the weight loss during the testing.
A slight reduction in rolling resistance may be observed, however the change is less than the
scatter of the data, on the order of 1 percent to 2 percent change for a 1-pound loss of tread com
pound.
73
Table 37 shows the ANOVA analysis of the data for rolling resistance loss versus weight loss
and heel-to-to wear value for the tires. Heel-to-toe wear is a more significant term than weight
loss, but the overall trend is still within the scatter of the data. Figure 49 shows the percentage of
the original rolling resistance versus the weight loss and heel-to-to wear. As expected from the
low R2 value of 0.13, the scatter in the data is too great to draw conclusions about the effects of
these changes on rolling resistance. All data is within 6 percent of the original rolling resistance,
so neither term has a significant effect on the rolling resistance.
RR, lbs
16
15
14
13
12
11
10
9
42 43 44 45 46 47 48 49 50 51 52
Tire Weight, lbs

Figure 47. ISO 28580 Rolling Resistance Versus Tire Weight Loss
74
RR, %
103
102
101
100
99
98
97
96
95
94
93
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Weight Loss, lbs

Figure 48. Rolling Resistance as Percent of the Original Rolling Resistance
Versus Tire Weight Loss During Testing
75
Table 37. Analysis of Rolling Resistance Change

Versus Weight Loss and Heel-to-Toe Wear
Dependent Variable: Rolling Resistance as a Percent of New Tire
Model 2 23.3716819 11.6858410 3.05 0.0581
Error 41 156.9477014 3.8279927
R-Square Coeff Var Root MSE rr Mean
0.129613 1.984430 1.956526 98.59384
Weight Loss 1 0.41847204 0.41847204 0.11 0.7426
Change in Heel-to-Toe Wear 1 6.00659293 6.00659293 1.57 0.2174
Intercept 99.37003782 0.43386711 229.03 <.0001
Weight Loss -0.35589632 1.07640530 -0.33 0.7426
Change in Heel-to-Toe Wear -1.07151522 0.85540108 -1.25 0.2174
76
RR, %
102.92
99.89
96.87
-0.37
-0.85
93.84
1.64 -1.32 ht
1.15
0.65 -1.80
Weight Loss, lbs 0.16
Figure 49. Percentage of Original Rolling Resistance
Versus Weight Loss and Heel-to-Toe Wear Change
77
4.0 CONCLUSIONS
Based on five different fuel economy cycles, a 10 percent decrease in tire rolling resistance re
sulted in approximately 1.1 percent increase in fuel economy for the vehicle. This result was
within the range predicted by technical literature. Reducing the inflation pressure by 25 percent
resulted in a small but statistically significant decrease of approximately 0.3 to 0.5 miles per gal
lon for four of the five fuel economy cycles, excluding the high-speed, high-acceleration US06
cycle. This value was smaller than many values predicted by technical literature, and possible
explanations are being explored.
For the tires studied, there appeared to be no significant relationship between dry peak or slide
numbers and rolling resistance. However, these tire models exhibited a strong and significant
relationship between better rolling resistance and poorer wet slide numbers. The peak wet slide
number displayed the same tendency, but the relationship was much weaker. This may be sig
nificant to consumers without anti-lock braking systems (ABS) on their vehicles since the wet
slide value relates most closely to locked-wheel emergency stops. For newer vehicles with ABS
or electronic stability control systems, which operate in the lower slip and higher range of wet
peak friction, the tradeoff is expected to be less significant.
For the subset of five tire models subjected to on-vehicle treadwear testing (UTQGS), no clear
relationship was exhibited between tread wear rate and rolling resistance levels. For the subset of
six tire models subjected to significant amounts of wear in the indoor treadwear tests, there was a
trend toward faster wear for tires with lower rolling resistance.
78
5.0 REQUIREMENTS
The Energy Independence and Security Act of 2007 specified in SEC. 111. Consumer Tire In
formation that the Secretary of Transportation shall, after notice and opportunity for comment,
promulgate rules establishing a national tire fuel efficiency consumer information program that
includes:
(A) a national tire fuel efficiency rating system for motor vehicle replacement tires to as
sist consumers in making more educated tire purchasing decisions;
(B) requirements for providing information to consumers, including information at the
point of sale and other potential information dissemination methods, including the Inter
net;
(C) specifications for test methods for manufacturers to use in assessing and rating tires
to avoid variation among test equipment and manufacturers; and
(D) a national tire maintenance consumer education program including, information on
tire inflation pressure, alignment, rotation, and tread wear to maximize fuel efficiency,
safety, and durability of replacement tires.
The recommendations of the technical staff will therefore be summarized per each specific re
quirement.
(A) A national tire fuel efficiency rating system for motor vehicle replacement tires to assist
consumers in making more educated tire purchasing decisions
Phase 1 of the project showed that the current laboratory rolling resistance test methods provide
an objective and repeatable basis for measuring the energy loss per unit distance traveled for a
given tire at a set of nominal operating conditions. This energy loss requires the vehicle to supply
additional torque to the rotating tire and directly reduces the efficiency of a vehicle in converting
the chemical energy in the fuel to motion of the vehicle. Therefore, tire rolling resistance is the
most effective metric for rating the “fuel efficiency” of a tire. 13
(B) Requirements for providing information to consumers, including information at the point
of sale and other potential information dissemination methods, including the Internet
The output of a laboratory rolling resistance test, either rolling force in units of N or lbf, or roll
ing resistance coefficient in N/kN, kg/tonne, or lbf/kip will be difficult for consumers to under
stand and relate to vehicle fuel economy. These measures must be communicated in a more con
sumer friendly format.
13
It should be noted that Schuring defines “tire efficiency” as the ratio of tire output to tire input energy. In the case
of a tire measured on a rolling resistance test machine, the “tire efficiency” of the free-rolling tire (zero slip - steady
state speed) is technically zero. Schuring, D. J., & Futamura, S. (1990). Rolling Loss of Pneumatic Highway Tires
in the Eighties. Figure 7. Rubber Chemistry and Technology, Vol. 62, No. 3, pp. 315–367.
79
(C) Specifications for test methods for manufacturers to use in assessing and rating tires to
avoid variation among test equipment and manufacturers
Phase 1 of the project showed that all five of the rolling resistance test methods evaluated had
very low variability and all methods could be cross-correlated to provide the same information
about individual tire types. It was concluded that while multi-point rolling resistance test meth
ods are necessary to characterize the response of a tire’s rolling resistance over a range of loads,
pressures, and/or speeds, either of the two shorter and less expensive single-point test methods
were sufficient for the purpose of simply assessing and rating individual tires in a common sys
tem. The single-point ISO 28580 was preferable to the single-point SAE J1269 method since the
former contains a lab-to-lab measurement result correlation procedure. In addition, the Commis
sion of the European Communities (EU) has selected ISO 28580 international standard as the
basis of their rolling resistance rating system. Use of ISO 28580 would allow international har
monization of U.S. and European test practices. Should the ISO 28580 final draft international
standard fail to be finalized in time for use in the system, or be cancelled, the SAE J1269 test is a
reasonable alternative. If the two ISO 28580 lab alignment tires fail to be made available, either
single-point rolling resistance test could use the ASTM F2493 16-inch standard reference test
tire for continued lab alignment (a concept proven in Phase 1), and perhaps the ASTM E1136
14-inch SRTT as a second tire.
The lab alignment procedure in ISO 28580 14 , which for passenger tires uses two dissimilar tires
to calibrate a test lab to a master lab, states that it will compensate for differences induced from
tests conducted using different options under the test standard. These options include the use of
one of four measurement methods (force, torque, power, or deceleration), textured or smooth
drum surface, correction of data to a 25C reference temperature, and correction of data from
tests conducted on a test drum of less than 2.0-m in diameter to a 2.0-m test drum. The variabil
ity in test results induced by allowing the various test options, as well as the effectiveness of the
temperature and test drum correction equations is not currently known to the agency. The ISO
TC 31 technical committee 15 responsible for the development of the 28580 rolling resistance
standard has 20 participating countries, of which the United States through ANSI 16 is the secre
tariat, and an additional 31 observing countries. Therefore, the ISO test standard strives to be
functional with the various technical capabilities present in the 51 member and observer coun
tries. When such a test standard is applied for regulatory use, in which compliance testing will be
14
Per ISO 28580, Section 10.2.2: “The reference machine laboratory control tyre monitoring must occur at a maxi
mum interval of one month. Monitoring must include a minimum of 3 separate measurements sometime during this
one month period. The average of the 3 measurements made during a one month interval must be evaluated for drift
from one monthly evaluation to another.’ Per ISO 28580, Section 10.5.5: “The alignment process must be repeated
at least every second year and always after any significant machine change or any drift in candidate machine control
tyre monitoring data.”
15
ISO TC 31 Tyres, rims and valves.
http://www.iso.org/iso/standards_development/technical_committees/list_of_iso_technical_committees/iso_technica
l_committee.htm?commid=47670
16
American National Standards Institute (ANSI)
http://www.iso.org/iso/about/iso_members/iso_member_body.htm?member_id=2188
80
conducted and civil penalties for non-compliance may be levied 17 , it may be desirable to limit
the application of the test method options so as to lessen the variability in the data submitted to
agency. This in turn may facilitate more accurate ratings, and more effective compliance and en
forcement activities.
The first option to consider is the type of measurement methods allowed for testing. The use of
force, torque, and power methods are permitted by the domestic SAE standards, with force and
torque being the most common. There is limited international use of the fourth method, decelera
tion, which is allowed under ISO test standards. The agency’s evaluation of the variability of the
ISO 28580 test in Phase 1 could only be carried out on the available test equipment in the United
States, which were 1.707-m test drums with force or torque measurement methods. Due to the
indirect nature of the power and deceleration measurement methods, as opposed to the more di
rect force or torque measurement methods, and the agency’s lack of access to test machines in
the United States that use power or deceleration methods, constraining the use of ISO 28580 to
only force or torque methods of measurement for data submission to the consumer information
system is advised.
A second option allowed under ISO 28580 is the use of the specified smooth steel or optional
textured (a.k.a. grit) material on the drum surface. Each surface has its own benefits and trade
offs when compared to the other, which mainly revolve around maintenance of the test surfaces.
The smooth steel test drum specified in ISO 28580 requires frequent cleaning to remove tread
rubber and oil build-up, which can alter test results. The roughness of the steel surface is to be
maintained (presumably through grinding or polishing) at maximum centerline average height
value of 6.3 μ-m. The approximately 1/16th to 1/8th-inch thick textured surface material (similar
to an 80-grit adhesive sand paper or 3M Safety-Walk tape) requires installation, conditioning 18 ,
and eventual replacement. However, the textured surface inhibits a build-up of tread rubber and
oil on the test drum surface. The master lab for lab alignments under ISO 28580 will use a 2.0-m
smooth steel roadwheel. It is not known whether the proposed regional labs used for lab align
ment will test on a smooth or a grit surface.
All rolling resistance methods utilize a skim test to determine the non-tire losses of the test
equipment, which for the force and torque methods includes bearing and aerodynamics losses.
Skim measurements are conducted by loading the tire to a value just sufficient to maintain tire
rotation at test speed without slippage. The ISO 28580 standard states that a textured surface may
be used to improve the skim test reading accuracy. The rolling resistance of a tire is known to
increase with increased surface roughness. LaClair cited research by Luchini (1983) indicating a
grit drum surface, which is intended to more closely mimic a road surface, is also more repeat
able than a smooth steel surface. However, a grit surface can generate rolling resistance numbers
2-11 percent higher than a smooth surface.[38],[39] In NHTSA Phase 1 testing, the rolling resis
tance of deep-lug tires exhibited a relatively linear behavior on grit surfaces over a range of test
17
Energy Independence Security Act(b) ENFORCEMENT.—Section 32308 of title 49, United States Code, is
amended—(c) SECTION 32304A.—Any person who fails to comply with the national tire fuel efficiency informa
tion program under section 32304A is liable to the United States Government for a civil penalty of not more than
$50,000 for each violation.’’
18
A conditioning procedure for a new textured (grit) surface is specified in SAE J2452: Issued JUN1999, Appendix
B, Surface Conditioning Procedure.
81
loads but dropped off at the lighter loads on smooth steel drums. This was attributed to slippage
of the deep lug tires on the smooth 19 surface. Deep lug passenger vehicle tires (many of which
are P-metric) are more common in the U.S. market, where textured drum surfaces are commonly
used, than in Europe where smooth drum surfaces are commonly used. Additionally, U.S. test
labs generally lack the capability to initially obtain and then maintain the surface roughness re
quirements of the smooth steel test drum. Conversely, the European labs have little experience
with the textured surface. In consideration of this, it is recommended that agency allow results
from both surfaces for data submission. However, agency compliance testing should be con
ducted on the more accurate textured surface used domestically, with lab alignment theoretically
providing the 2-11 percent correction to a smooth surface.
A third option allowed under ISO 28580 is the temperature correction of data to 25C from tem
peratures within the 20-30C range. It is known that rolling resistance varies with temperature
during on-vehicle operation. According to LaClair, “The variation in rolling resistance as a
function of temperature is not linear. However, between 10 and 40C, an increase of 1C corre-
sponds to a reduction in rolling resistance of about 0.6 percent under normal road opera-
tion.”[38] Assuming this relationship applies to laboratory testing as well, the 5C (9 F) range
allowed by ISO 28580 may translate into a maximum of 3% variation in rolling resistance.
Within the range of permissible laboratory ambient temperatures, the standard specified a linear
formula that corrects the Fr value of passenger tires by ±0.8 percent for each degree Celsius the
temperature departs from 25C. The correction is plotted in Figure 50. For passenger tires, this
correction can reach a maximum of 4 percent of the measured rolling resistance force. There
fore, the correction appears to be reasonable in magnitude when compared to the stated road op
eration response of rolling resistance to temperature.
19
While the likely explanation, the test lab was not equipped to measure tire slip during operation. It is also signifi
cant to note that the “bare steel” wheel used in the NHTSA testing did not have a finish certified to average height of
6.3 μ-m.
82
1.06
1.05
1.04
Temperature Correction Factor (ISO 28585) Truck and bus (load index < 121)
1.02
Passenger
0.98
0.96
0.95
0.94
20 21 22 23 24 25 26 27 28 29 30
Ambiant Temperature (Celsius)
Figure 50. Temperature Correction Factor - ISO 28580
A fourth option allowed under ISO 28580 is the correction of data to 2.0 meter drums commonly
used in Europe from smaller test drums, such as the 1.707-m (67.23-in.) test drums commonly
used in the United States. Since no 2.0-meter drums were available for testing in the United
States, the variation in test results resulting from use of this formula have not been verified by
the agency. Figure 51 shows the equation in ISO 28580 used to correct Fr measured on a 1.707
m drum to a 2.0-meter drum equivalent over a range of tire radii from 0.3 m to 0.5 m. The cor
rection factor varies from 0.98 to 0.97 over this range, or 2 to 3 percent of measured rolling
force. The correction equation is based on a theoretical concept that was not validated by the
agency with different tire designs and sizes. This again may introduce variability in the data re
ported to the system. At a minimum, reporting of results in terms of a single drum diameter will
be necessary to prevent rating compliance disputes.
83
0.984
0.982 Correction Factor from 1.707 m to 2.0 m drum diameter
0.98
0.978
Correction Factor
0.976
0.974
0.972
0.97
0.968
0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
Tire Radius (m)
Figure 51. Drum Diameter Correction Factor - ISO 28580
(D) A national tire maintenance consumer education program including, information on tire
inflation pressure, alignment, rotation, and tread wear to maximize fuel efficiency, safety, and
durability of replacement tires.
The agency’s www.Safercar.gov website contains information and recommendations on tire in
flation pressure, alignment, rotation, and treadwear. The test results from Phase 2 allow estimates
of the effects of tire rolling resistance levels on fuel efficiency, safety, and the durability of tread
of light vehicle replacement tires. The results also attempt to quantify the effects of tire underin
flation on vehicle fuel economy. These results could be cited in consumer information.
84
6.0 ROLLING RESISTANCE (FR) VERSUS ROLLING RESISTANCE COEFICIENT

CR)
An important facet of the rating system is data reporting. The two candidates for reporting of the
data under the draft ISO 28580 standard are the rolling resistance (Fr) and the rolling resistance
coefficient (Cr). The ISO 28580 standard defines rolling resistance as the “Loss of energy (or
energy consumed) per unit of distance travelled.” The standard defines rolling resistance coeffi
cient (Cr) as the “Ratio of the rolling resistance, in newtons, to the load on the tire, in knewtons.
This quantity is dimensionless.” [40]
6.1 Theory of Fr and Cr

Schuring and Futamura define the units of rolling resistance as watt-seconds per meter, or joules
per meter (J/m), which is equivalent to newton meters per meter (N-m/m).[41] The meter per
meter (m/m) is sometimes cancelled out in this notation to provide Fr in terms of newtons (N)
alone. However, Schuring and Futamura emphasize that while the unit J/m is dimensionally
equivalent to the newton (N), rolling resistance is not a force but rather an energy loss per dis
tance traveled, which is a scalar with no direction (i.e., not a vector).[41],[42] Therefore, ISO
28580 and ISO 18164 note that “The SI unit conventionally used for the rolling resistance is the
newton metre per metre (N m/m). This is equivalent to a drag force in newtons (N).”[43] With
the exception of SAE J1269 (circa 1979) 20 , the three most recent standards, J2452 (circa 1999),
ISO 18164 (circa 2005) 21 , and the draft ISO 28580 (expected 2009) have adopted language de
fining rolling resistance as the energy a tire consumes per unit distance (See Definitions section
of this report).
In ISO 28580, a laboratory rolling resistance test machine may use the force, torque, power, or
deceleration method to calculate the rolling resistance at interface of the tire and drum. Subtrac
tion of skim values and corrections are conducted with the data in the Fr format, then the rolling
resistance coefficient (Cr) is determined by dividing the Fr by the specified test load on the tire.
The concept of rolling resistance coefficient (Cr) stems from the fact that, ignoring vehicle in
puts, the equilibrium (i.e., fully warmed up) rolling resistance of a new radial tire varies primar
ily with applied load, inflation pressure, and speed. Investigations such as those by Clark et al.
during the 1970s indicated that the equilibrium tire rolling resistance of radial passenger tires
was not linear with pressure or speed, but did appear linear with load. In the 1979 handbook pre
pared for the U.S. Department of Transportation, Clark and Dodge explain the concept and ap
plication of the rolling resistance coefficient[44]:
“In all four of these sets of data (two bias and two radial 22 tire models) the linear rela
tionship between load and rolling resistance is very close, and further, to a very close ap
proximation the rolling resistance vanishes at zero load, with a straight line drawn
through the data points nearly intersecting the origin of rolling resistance and load. …
20
SAE J1269 was originally issued in 1979, reaffirmed in 2000, and revised and reissued in 2006.
21
ISO 18164 was issued in 2005 but states that it is a compilation of three older individual standards (ISO
8767:1992, ISO 9948:1992, and ISO 13327:1998) that have since been withdrawn.
22
Goodyear GR 78-14 and Uniroyal HR78-15.
85
The linear nature of the equilibrium rolling resistance as a function of load is apparently
fortuitous, but is well known and has led to the common and very useful concept of the
coefficient of rolling resistance, which is defined as the rolling resistance divided by the
load carried.”
In their paper, the authors continue on to explain how the rolling resistance coefficient can be
used to evaluate different tires for a known vehicle[45]:
“The coefficient of rolling resistance is a convenient concept since it allows one to com
pare various tires for use on the same vehicle. The load carried by a tire will be the same
on a given vehicle in a given tire position, so a comparison of the rolling resistance coef
ficients will show which tire is the most efficient for a given application. On the other
hand, tests of tire rolling resistance are usually carried out at the tire rated load or at some
relatively large fraction of it, such as 80 percent of tire rated load. Direct presentation of
the rolling resistance under these conditions is dependent on the load carried by the tire,
which, of course, varies for different tire sizes. Hence, the concept of the coefficient is a
generalizing and extremely useful one for both the presentation and interpretation of
data.”
Therefore, the concept of rolling resistance coefficient (Cr) would appear advantageous when
calculating the expected rolling resistance of a tire, or of tires of different load ranges or sizes,
for a “given application” (i.e., for a given vehicle with known wheel corner loads). The coeffi
cient Cr transforms the “energy per unit distance” measure of Fr into terms of “energy per unit
distance and unit load” on the tire. As stated earlier, no simple relationship exists between roll
ing resistance and pressure or speed that would allow the calculation of similar coefficients for
these two inputs.
To determine the sensitivity of a tire’s rolling resistance to load and pressure, the first rolling re
sistance test standard, SAE J1269 (1979), evaluated tire rolling resistance over a range of three
pressures and two loads at 80 km/h (50 mph) (Figure 52). For passenger tires, the two test loads
are 50 and 90 percent 23 of the maximum load limit of the tire. The combination of pressure and
load conditions result in four discrete test points (TP 1 to TP 4). Skim loads are subtracted from
each test point and the data is corrected while still in terms of Fr. If desired, the standard speci
fies an option to fit a least-squares regression model to the data, which uses separate equations
for passenger, light truck, and highway truck and bus tires. The linear regression equation for
passenger car tires is:
23
90 percent of maximum rated tire load is a logical upper limit for test load, since FMVSS 571.110 requires that
the vehicle normal load on a tire not exceed 94 percent of the rated load of the tire at the vehicle manufacturer’s rec
ommended cold inflation pressure of the tire. For passenger tires installed on MPV, truck, bus, or trailers, the allow
able rated load of the tire is reduced by 10 percent and the normal load must still not exceed the 94 percent of the de-
rated load.
86
FR = FZ(A0+A1FZ+A2/p)
FZ = Tire load (N [lbf])
p = Equilibrium inflation pressure (kPa [psi])
A0, A1, A2 = Coefficients
Equation 4. SAE J1269 Linear Regression Equation for Passenger Car Tires
Test Speed of 80 km/h [50 mph]

Pressure
Pr+70 kPa TP 4 TP 2
Pr -30 kPa TP 3
Pr -50 kPa capped TP 1
40% 50% 60% 70% 90%

Load
Figure 52. SAE J1269 Recommended Test - Evaluates Response of Rolling Resistance
Force Over a Range of Three Pressures and Two Loads
After determining the coefficients of the equation in J1269, a predicted rolling resistance can be
calculated at any load and pressure. 24 In the original SAE J1269, the Cr is determined by divid
ing the Fr by the corresponding test load on the tire. Since Cr is assumed to be a constant, any Fr,
whether measured or predicted by the regression equation, can be used in the calculation. The
latest version of SAE J1269 (2006) specifies a Standard Reference Condition (SRC), consisting
of a single load and pressure, from which Equation 4 can be used to calculate a standard Fr and
Cr. This latest version of the standard still recommends use of the multi-point test, but states that
the test may be conducted at the single-point SRC conditions “which may be used for the pur-
pose of high volume comparisons.”[46] However, no version of J1269 states how Cr, whether
determined from multi or single-point methods, is to be used.
ISO 18164 (1992-1998) 25 specifies a rolling resistance test with a single load and single inflation
condition, which can be run at either a single speed or three speeds. Annex B of the standard
specifies optional test conditions for determining the speed and/or load and inflation sensitivity
of a tire. The standard states[43]:
24
SAE J1269 (SEP, 2000, Sept.) p. 10 states: “The resulting regression equation may be used to calculate values for
rolling resistance at loads and pressures other than those tested, but extrapolation far beyond the range of the test
matrix, particularly for the region of high load and low pressure, is not advised.”
25
ISO 18164 was issued in 2005 but states that it is a compilation of three older individual standards (ISO
8767:1992, ISO 9948:1992 and ISO 13327:1998), which have since been withdrawn.
87
“The rolling resistance of a tyre will vary with speed, load and inflation pressure, as well
as other factors. Depending on the circumstances of particular tyre applications, it can be
useful to determine the effect of these tyre-related parameters for the individual tyre to be
tested. If such information is desired, the options indicated in (Annex) B.2 and B.3 are
recommended.”
In Annex B.2 of ISO 18164, the speed sensitivity of passenger tires is evaluated at 50 km/h, 90
km/h and 120 km/h in sequence. In Annex B.3, the load and inflation sensitivity of passenger
tires are evaluated at two loads, 50 and 90 percent of maximum load, and two pressures, +70 kPa
and -30 kPa from the single-point pressure (Figure 53). Like the preceding SAE J1269, ISO
18164 subtracts skim loads and corrects the data in terms of Fr. Unlike J1269, 18164 does not
contain an option in Annex B to fit a regression equation to data from multiple loads and pres
sures. If using the multi-point test conditions, a Cr must be determined from dividing a measured
Fr by its corresponding test load. Again, since Cr is assumed to be a constant, any measured Fr
can be used in the calculation. The ISO 18164 standard also does not state how Cr is to be used.
Pressure Test Speeds of 50 km/h, 90 km/h and 120 km/h
Pr+70 kPa TP 1 TP 3
Pr -30 kPa TP 2 TP 4
50% 90%
Load
Figure 53. ISO 18164 Annex B - Response of Rolling Resistance Force (Fr) Over a Range of
Three Speeds, Two Pressures, and Two Loads
The later SAE J2452 (circa 1999) goes farther in continuously measuring rolling resistance over
a stepwise speed coastdown from 115 to 15 km/h (71 to 9 mph). As with SAE J1269 and ISO
18164, J2454 recommends testing at a matrix of loads and pressures[47]:
“In order to obtain a complete quantification of tire rolling resistance as a function of

load, inflation pressure, and speed, the load/pressure matrices specified in 7.2.1 should be
88
used. However, if needed, the stepwise coastdown can be performed for a single
load/pressure condition.”
The first data reduction process uses a mathematical model to describe a tire’s rolling resistance
as a function of load, inflation pressure, and speed. Interestingly, while the J2452 test includes a
definition of Cr, it does not calculate Cr in the standard. Instead, the standard calculates a mean
equivalent rolling force (MERF), which is the average rolling resistance of a tire at a
load/inflation condition over a driving cycle with a specified speed-time profile. J2452 also al
lows calculation of a standard mean equivalent rolling force (SMERF) at a single-point reference
condition (a single load, pressure, and speed).
To save time and expense, the draft ISO 28580 rolling resistance standard calculates rolling re
sistance Fr at single load, pressure, and speed (Figure 54). Subtraction of skim values and correc
tions are conducted with the data in the Fr format, then the rolling resistance coefficient (Cr) is
determined by dividing the Fr by the nominal test load on the tire (Equation 5).
Cr = Fr/Lm
Cr = Rolling resistance coefficient (dimensionless)

Fr = Rolling resistance in newtons
Lm = Test load in knewtons
Equation 5. ISO 28580 Rolling Resistance Coefficient
Pressure Test Speed of 80 km/h (50 mph)
210 kPa capped TP
80%
Load
Figure 54. ISO 28580 Test Conditions for Standard Load Passenger Tires
As with the three other test standards, there is no mention in ISO 28580 of how Cr is to be used.
However, the test standard states in its scope[40]:
89
“Measurement of tyres using this method enables comparisons to be made between the
rolling resistance of new test tyres when they are free-rolling straight ahead, in a position
perpendicular to the drum outer surface, and in steady-state conditions.”
The most straightforward interpretation is that the rolling resistance coefficient in ISO 28580 is
intended to normalize rolling resistance by test load to allow a relative comparison of the energy
consumption of tires of all sizes and load ranges. However, the previous discussion has illus
trated how the Cr coefficients from multi-point (multi-load) rolling resistance are used to calcu
late the rolling resistance of a tire at a known wheel load (vehicle load divided by four), usually
for the purpose of evaluating a tire or tires for a given vehicle. This calls into question whether
the Cr calculated from a test at single load can also be used for such purposes.
6.1.1 Using Cr from a Single-Load Test to Predict Rolling Resistance at Any Load
There are a number of assumptions that must be fulfilled to be able to predict the response of a
tire’s rolling resistance over a range of loads from measurement of rolling resistance at a single
load. First, since a single-point in space can have an infinite number of lines pass through it, a
second point must be defined in order to determine the sensitivity of a tire’s rolling resistance to
load. For the purposes of a single-point Cr, this second point is defined as the origin (Figure 55).
Since this function is a straight line defined by two points, the actual response of rolling resis
tance to load changes should be fairly linear or errors will be induced. Second, to use Cr as a sca
lar to vehicle load, the rolling resistance coefficient should be constant (i.e., a flat line) over the
range of practical tire loads or errors will be induced (Figure 56).
Test Speed of 80 km/h (50 mph),

Capped Pressure of 210 kPa (30 psi)
Tire A
Theoretical Fr
Rolling Resistance (N)
Tire B
80%
Load
Figure 55. Theoretical Single-Load Rolling Resistance (Fr)
90
Test Speed of 80 km/h (50 mph),

Capped Pressure of 210 kPa (30 psi)
Theoretical Cr
Tire A
Rolling Resistance
Coefficient
(Dimensionless) Tire B
80%
Load
Figure 56. Theoretical Single-Load Rolling Resistance Coefficient (Cr)
In Phase 1 of this project, the agency measured the rolling resistance of 16 passenger tire models
in a number of single and multi-point tests. Figure 57 displays rolling resistance data for the tires
over a range of loads in the various tests (all points were collected at the identical pressure and
speed). Note that the two points are connected with straight lines to emphasize that the Fr is not a
linear function passing through the intercept. It is likely that the actual Fr values do pass through
the intercept (i.e., there is zero rolling resistance at zero load), but that the function is actually
non-linear as is hypothesized in the SAE J1269 (multi-point) regression shown in Equation 4.
Figure 58 displays rolling resistance (Cr) data for same tires over the range of loads. It’s impor
tant to note that the Cr values in Figure 58 at different loads are not constant, sometimes increas
ing and sometimes decreasing with load depending on the given tire model. In other words, Cr
does not appear to be a constant coefficient, which is why the multi-point tests evaluate rolling
resistance over a range of loads and use non-linear regressions to predict a tire’s response to
load.
91
Fr
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
0 200 400 600 800 1000 1200 1400 1600
Load
Type B10 B11 B12 B13 B14 B15 D10 G10
G11 G8 G9 M13 M14 P5 R4 U3
Figure 57. Rolling Resistance of 16 Passenger Tires
Versus Load at Constant Pressure (Average of 8 Values)
92
Cr
0.012
0.010
0.008
0.006
600 800 1000 1200 1400 1600
Load
Type B10 B11 B12 B13 B14 B15 D10 G10
G11 G8 G9 M13 M14 P5 R4 U3
Figure 58. Rolling Resistance Coefficient of 16 Passenger Tires
Versus Load at Constant Pressure (Average of 8 Values)
Beyond the inconsistencies with Cr, there exist practical problems in that very few vehicles are
operated at the GAWR/GVWR listed on the placard, and few tire dealers have vehicle scales that
allow determination of actual vehicle weight. Without a know corner load for a tire, the Cr can
not be used to calculate a rolling resistance for a given tire model. A standard estimate of per
centage of a vehicle’s GVWR to use Cr to estimate Fr would likely not be more predictive than
the Fr measured at 80 percent of maximum tire load rating. Also, there comes additional diffi
culty in predicting the rolling resistance of a tire for a given vehicle from a single-pressure test.
The allowable placard inflation pressures for standard load passenger car tires range from 180
kPa (26 psi) to 240kPa (35 psi), and up to 280 kPa (41 psi) for extra load tires. No similar coeffi
cient is available from ISO 28580 26 to correct the expected Fr from the 210 kPa (30 psi) standard
load (250 kPa [36 psi] extra load) test pressure in the standard to the actual placard operating
pressure of the vehicle, which can differ by axle. Therefore, the idea of calculating rolling resis
tance for a specific vehicle is not usually possible with Cr, unless its tires operate at the ISO
26
Note that the coefficient in the SAE J1269 test for passenger tires is A2/p: rolling resistance varies by the inverse
of the inflation pressure.
93
28580 test pressure, or a multi-point rolling resistance test is used to generate a regression equa
tion from tests at multiple pressures.
6.2 Discussion
It has been asserted that Cr would be more useful than Fr as a basis of rating tires for consumers
who are looking to replace tires on their vehicle with tires of the same size but different maxi
mum load ratings. The FMVSS No. 139 allows tire maximum load ratings to be determined from
one of six international organizations, 27 or to be specified to the agency by an individual manu
facturer. For example, the agency’s Phase 1 research used a large number of tire models of the
most popular P-metric replacement tire size in 2007, which was P225/60R16. The standard load
P225/60R16 Goodyear Integrity tire (type G12), which was OE on the test vehicle, has a load
index of 97 that allows it to carry a maximum of 730 kg (1609 lbs) at maximum pressure. The
metric designated 225/60R16 Goodyear Integrity tire (type G8) has a load index of 98, allowing
it to carry 750 kg (1653 lbs), or 20 kg (44 lbs) more at maximum pressure. Per ISO 28580, both
tires are tested at 80 percent of maximum load, resulting in the G8 tire being tested at 16 kg (35
lbf) more load in the rolling resistance test. In this test, the average rolling resistance of the
P225/60R16 Integrity tire was 9.47 lbs, and the 225/60R16 was 9.83 lbs, a 0.36 lbf (+4%) differ
ence.
To adjust for the different test loads, the rolling resistance coefficient (Cr) is calculated. Ac
counting for significant digits, the Cr of the P225/60R16 is 9.47 lbf / 1287 lbf = 0.0074 lbf/lbf
and the Cr of the 225/60R16 is 9.83 lbf / 1322 lbf = 0.0074 lbf/lbf. Therefore, since the Cr values
were identical, the 4 percent difference between the two Integrity tires likely resulted from the
different test loads, not the tires themselves. If the tires were rated strictly on the ISO 28580 Fr
magnitudes, the P225/60R16 tire has lower rolling resistance than the 225/60R16 tire. This issue
has implications in that for many sizes of tires, the metric designated tires (usually of European
or Asian manufacture) have a marginally higher load index than the P-metric tires. 28 As a result,
the metric tires would be tested at higher loads than P-metric tires of the same size and yield
slightly higher rolling resistance. However, this does not appear to be a penalty in that a tire of a
given size that is rated with a higher load index, for instance a 98 load index rather than a 97,
could be operated at higher loads on heavier vehicles and actually generate more rolling resis
tance.
27
The tire load rating shall be that specified either in a submission made by an individual manufacturer, pursuant to
S4, or in one of the publications described in S4 for its size designation, type and each appropriate inflation pressure.
If the maximum load rating for a particular tire size is shown in more than one of the publications described in S4,
each tire of that size designation shall have a maximum load rating that is not less than the published maximum load
rating, or if there are differing maximum load ratings for the same tire size designation, not less then the lowest pub
lished maximum load rating. S4 (1) The Tire and Rim Association; (2) The European Tyre and Rim Technical Or
ganization; (3) Japan Automobile Tire Manufacturers’ Association, Inc.; (4) Tyre & Rim Association of Australia;
(5) Associacao Latino Americana de Pneus e Aros (Brazil); (6) South African Bureau of Standards. (Source:
FMVSS No. 571.139.
28
In a survey of 69 tire sizes sold by the Tire Rack in both P-metric and Euro-metric sizes: 12 percent had equal
load designations, 85 percent had load designations from 1 to 6 load index numbers higher (average of 1.5) for the
Euro-metric size and 1 size had a higher load index designation for the P-metric tire.
94
Nonetheless, normalizing all tires to their test load with Cr in order to provide a relative measure
of their rolling resistance may be useful if the normalization is indeed consistent across all tire
sizes. It is therefore necessary to think outside the context of selecting tires for a known vehicle
and consider the rating system as a whole. Neither Fr nor Cr have been used before to rate a large
population of tires in a common rating system. It is absolutely factual to state that for a given ve
hicle, which has a single nominal tire load, Fr and Cr will produce identical rankings of tires of
the same size and load index. However, the proposed tire fuel efficiency rating system must rate
all tires in the system independently of specific vehicles, and recognize that a given tire model
may be operated at many different loads. In 2009, Lambillotte estimated that a rolling resistance
rating system in the United States may cover greater than 20,000 individual passenger tire stock-
keeping units (i.e., unique tire brand/model/size/pattern, etc., designations).[48] Therefore, it is
important to consider the implications of using Cr to categorize a wide range of tires in a rating
system. When Cr is applied over a large range of tire sizes, it tends to produce lower relative val
ues for larger tires than for smaller tires, despite the fact that the larger tires will very likely use
more energy. This in turn skews the grades of tires when compared in a common system. Schur
ing and Futamura reported this trend in 1980’s era tires (13-15 inch tires sizes)[49]:
“If a family of tires of different sizes would be tested for rolling loss at a maximum load
(prescribed by the Tire and Rim Association), or at a fixed fraction of maximum load, as
well as at a constant pressure and constant speed, and if rolling loss would be directly
proportional to maximum load (or a fraction thereof), then by definition, the rolling loss
coefficient derived from these test would be independent of size. This however is not the
rule. Rolling loss does increase not quite in proportion with increasing maximum load (or
fractions of it); hence, the rolling-loss coefficient of larger tires is mostly smaller than
those of smaller tires. … The reason for the slight decline in the rolling-loss coefficient
with tire size is not clear. We may speculate that the load formula (a rather complex em
pirical relation between permissible tire load, pressure, and tire dimensions, developed
and continuously amended over the decades by the Tire and Rim Association) had been
adjusted such that larger tires experience slightly lower strains than smaller tires.”
What Schuring and Futamura observed in 13- to 15-inch diameter tire sizes, and has since been
magnified as tires reach 30-inch diameters and beyond, is a result of the load term (Lm) in the
denominator of the Cr equation (Cr = Fr/Lm). This is where the non-linear formulas that deter
mine the maximum load ratings for tires have a large effect. For instance, Equation 6 in
Appendix 1 is the maximum load formula used by the Tire and Rim Association, Inc. Note the
multiple coefficients raised to powers, as well as the three different values for the K coefficient
depending on the aspect ratio of the tire. It is obvious that the Tire and Rim Association load
formula is going to provide three different, non-linear curves for maximum load across the range
of passenger tire sizes to be rated in the tire fuel economy system. Dividing the rolling resistance
force (Fr) by this non-linear and discontinuous function will result in a non-linear and discon
tinuous set of values for Cr. Additionally, certain P-metric tires of aspect ratios 30-45 have
maximum loads that do not follow the T&RA formulas, and were instead set equal to ISO loads
in order to harmonize internationally. Worse yet, a sizable portion of tires sold in the United
States are metric tires (tire sizes lacking a “P” at the beginning), and are rated by a different set
of equations under the ISO standards. The Tire Rack has an excellent description of the two sys
tems in layman’s terms[50]:
95
“P-metric sized tires are the ones with the "P" at the beginning of the tire size, (such as
P225/60R16 listed above). They were introduced in the United States in the late 70s and
are installed on vehicles primarily used to carry passengers including cars, station wag
ons, sport utility vehicles and even light duty pickup trucks. Their load capacity is based
on an engineering formula which takes into account their physical size (the volume of
space for air inside the tire) and the amount of air pressure (how tightly the air molecules
are compressed). Since all P-metric sizes are all based on the formula for load, vehicle
manufacturers can design their new vehicles (weights and wheel well dimensions) around
either existing or new tire sizes.
Metric or Euro metric sized tires are the ones without the "P" at the beginning, (such as
185R14 or the 225/60R16 listed above). Using metric dimensions to reflect a tire's width
actually began in Europe in the late 60s. However, since Euro metric sizes have been
added over time based on the load and dimensional requirements of new vehicles, the tire
manufacturers designed many new tire sizes and load capacities around the needs of new
vehicles. Not quite as uniform as creating sizes using a formula, but they got the job
done.”
Therefore, the idea of generating a linear, dimensionless coefficient in Cr by dividing Fr by 80

percent of maximum rated tire load puts either the three different T&RA non-linear load formu
las, or the ad hoc European system of load capacities into the denominator of the equation. While
the effects on selecting tires for a given vehicle are almost certainly negligible, the effects on rat
ing all tires of all sizes in a common system with Cr may be significant. For instance, Figure 59
shows the Fr calculated for values of passenger tire rolling resistance reported by the Rubber
Manufacturers Association (RMA) to the California Energy Commission [51] versus the load
index reported for the tires. Excluding what appear to be outliers, the values range from 5 pounds
to approximately 22 pounds. Figure 60 shows the values of Cr for the same tires. Excluding the
same tires that appear to be outliers, the values range from 6 to approximately 14. Two important
conclusions can be seen in this data:
1. The range of Fr values from lowest to highest is ~1.3 times the mean value for all tires,
while the range for Cr values is only ~0.8 times the mean value. This means that Fr will
have a greater ability to discriminate tires across the entire range of passenger tires. (As
previously noted, at a given load index the values for Fr and Cr are related by a constant
therefore the ability to discriminate tires at the same load index is identical.)
2. The average value for Fr increases with load index, meaning the amount of energy loss
(vehicle fuel consumption) is increasing as tire load indexes increase. However, the av
erage value for Cr decreases as tire load index increases. In fact, dividing by load does
not produce a “corrected” value for a tire that is independent of load, but rather a value
that is inverse to load.
96
Fr
35
30
25
20
15
10
75 85 95 105 115 125
Load Index
Figure 59: Rolling Resistance Force (SAE J1269 Single-Point, Pounds)
Versus Load Index for a Broad Range of Passenger Tires
Cr
18
17
16
15
14
13
12
11
10
9
8
7
6
75 85 95 105 115 125
Load Index
Figure 60: Rolling Resistance Coefficient (SAE J1269)
Versus Load Index for a Broad Range of Passenger Tires
97
This is where the goals of the fuel efficiency rating system must be considered. First and fore
most, the system should be intuitive to consumers. Consumers will use the system to purchase
tires for their current vehicle, as well as for subsequent vehicles, thus building up a contextual
understanding of the ratings over time. Also, consumers may have multiple vehicles in their
household or commercial fleet for which they purchase tires. A system based on the rolling resis
tance of each tire is directly relatable to fuel economy calculations and does not skew lar
ger/higher load tires into better ratings, such as a system using Cr as a basis. Regardless of
whether any two tire sizes in the system actually fit on the same vehicle, consumers could be
confused by a fuel efficiency system that gives equal or better ratings to larger tires that consume
more fuel than to smaller tires that consume less fuel.
For instance, in rating light vehicle fuel economy, the estimated fuel mileage given to consumers
is not divided by the rated payload capacity of the vehicle. Vehicle fuel economy ratings are in
stead an estimate of fuel efficiency of all vehicles in the system under the same set of driving
conditions. Given vehicle fuel economy, the consumer may then weigh the fuel efficiency of the
vehicle against any consideration such as payload capacity, top speed, number of occupant seats,
etc. Consumers who require certain cargo or towing capacities are no more able to choose a
smaller, more fuel efficient vehicle any more than a consumer with a large truck can choose a
small, low-rolling resistance tire. However, the estimated fuel economy of the light vehicles is
reported on the same basis regardless of vehicle type. Consumers should understand that heaviest
passenger vehicles tend to get the poorest fuel economy in part because the large tires operating
under the heavy loads of those vehicles consume more energy.
Another model is the UTQGS system. The UTQGS treadwear rating is intuitive to consumers in
that tires with higher grades will, under the same conditions, be expected to last longer than tires
with lower grades. This property is reported independent of any other tire property. Take for in
stance the speed category (maximum speed rating) of the tire. High-performance ZR, V, W, and
Y rated tires, which have much lower average treadwear grades than all season S, T, U, and H
rated tires, do not use a different reference tire for treadwear grading. Nor is the treadwear rated
divided by the speed category. Instead, all tires in the system are referenced on the same scale,
even though an ultra-high performance summer tire is likely not available in the OE sizes of a
minivan or economy car. The same is true of the traction and temperature resistance ratings. We
believer consumers expect high performance tires to have higher traction and temperature resis
tance ratings than S-rated tires, and would find a relative system, one in which a W-rated tire that
is expected to wear out in fewer miles is given a higher rating than an S-rated tire that is ex
pected to last longer, to be confusing.
An additional argument has been put forth that by providing consumers with fuel economy rec
ommendations for small and large tires on the same scale (use of Fr), rather than normalizing
everything to load capacity (use of Cr), the system may encourage consumers to choose smaller
tires with insufficient load carrying capacity for their vehicles, thus creating a safety hazard. This
rationale is flawed for many reasons. First, consumers have had a strong economic benefit to
purchase under-capacity tires for many decades, namely initial purchase price. The smaller tires
in a tire line normally cost less, and purchasing under-capacity tires would be an immediate eco
nomic benefit at the time of sale. This is contrasted with a future benefit of 6 to 12 gallons in an
nual fuel savings from purchasing tires with 10 percent lower rolling resistance than their current
98
tires.[52] The issue of lower-cost small tires has not manifested itself as a safety problem due
mainly to the fact that consumers lack the equipment to mount their own tires, and that tire in
stallers will not assume the legal liability for installing tires with insufficient load carrying capac
ity.
Finally, there comes the matter of calculating fuel economy from the output of the rolling resis
tance test. The calculated rolling resistance can be used to estimate a tire’s power consumption,
or when set equivalent to a drag force on a vehicle to calculate its impact on vehicle fuel con
sumption. The various analyses range from simple to highly complex fuel economy models. In
an example of a simplified approach, Pillai defined tire energy loss per hour “E(R)” equal to the
rolling resistance x distance traveled per hour.[53] For example, at the ISO 28580 test speed of
80km/h, a tire with a Fr of 50 N (50 N-m/m) consumes 1.1 kW of power per hour (50 N-m/m *
80 km/h * (1,000 m / 1 km) * (1 h / 3600 s) = 1111 N-m/s = 1.1 kW). For a tire with an Fr of 40
N, it consumes 0.8 kW of power per hour at 80 km/h. Therefore, rolling resistance (Fr) is a ratio
of the energy consumed per unit distance, which when expressed at a given speed can differenti
ate tires on the basis of expected power consumption.
The tire energy consumption or vehicle fuel economy approaches require rolling resistance in
terms of force for the calculations. Tires of vastly different drag forces can have identical rolling
resistance coefficients. Therefore, when the data is reported in terms of Cr, the coefficient must
be used to calculate an Fr at a known tire load, or the initial step of converting Fr to Cr at 80 per
cent of maximum tire load must be reversed. In other words, data reported in terms of Fr is di
rectly relatable to vehicle fuel economy. Whereas data reported in Cr must be transformed back
to Fr to allow vehicle fuel economy calculations. Given the nature of Cr to skew tires that con
sume more fuel into better relative ratings, the question persists as to the value of the extra step
of computing a single-point coefficient rather than reporting the data in terms of Fr.
99
Appendix 1. Tire and Rim Association, Inc. - Maximum Load Formula for “P” Type
Tires
Maximum Load “L” (kg) = (K) x (P0.50) x (Sd1.39) x (Dr + Sd) [54]
Variable 30 Series Through 35 Series 40 Series Through 45 Series 50 Series Through 80 Series
K 5.00 x 10-5 5.67 x 10-5 6.67 x 10-5
Sd [0.34848+0.6497(A)] x S.85 [0.34848+0.6497(A)] x S.70
A H/S.85 H/S.70
S.70 / S.85 Nominal Tire Section (mm)
H Section Height (mm)
Dr Rim Diameter Code (mm)
P Inflation Pressure (kPa); 240 kPa for Standard Load Tires or 280 kPa for Extra Load Tires
Equation 6. T&RA Load Formula for “P” Type Tires (S.I. Units)
100
Appendix 2. Detailed Test Matrix

Tire 1st Task 2nd Task EPA 5-Cycle Vehicle Fuel Indoor Tread- Outdoor Dry Outdoor Tread- Outdoor Wet Grand
Model Economy wear Traction wear Traction Total
Transportation Research Smithers Sci NHTSA San An NHTSA San NHTSA San An
Center, Inc. entific Ser gelo Test Facility Angelo Test gelo Test Facility
vices, Inc. Facility
B10 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
B11 J2452 3 1 1 5
J1269 - Single- J1269 - Single- 1 1
point point
J1269 - Multi 3 1 1 1 6
point
ISO 28580 1 1
ISO 18164 1 1
B12 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
B13 J2452 3 1 1 5
point point
J1269 - Multi 3 1 1 1 6
point
ISO 28580 1 1
ISO 18164 1 1
B14 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
B15 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
D10 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
G10 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
G11 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
G12 ISO 28580 ISO 18164 2 2
EPA 2008 TRC EPA 2008 TRC 2 2
G8 J2452 3 1 1 5
point point
J1269 - Multi 3 1 1 1 6
point
101
ISO 28580 1 1
ISO 18164 1 1
G9 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
M13 J2452 3 1 1 5
point point
J1269 - Multi 3 1 1 1 6
point
ISO 28580 2 2
M14 J2452 3 1 1 5
J1269 - Single- 1 1
point J1269 - Single- 1 1
point
J1269 - Multi 3 1 1 1 6
point
ISO 28580 1 1
P5 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
R4 J2452 3 1 1 5
J1269 - Multi 3 1 1 5
point
U3 J2452 3 1 4
J1269 - Multi 1 1 1 3
point
ISO 28580 2 1 3
Grand 52 48 32 20 32 184
Total
102
Appendix 3. Examples of Data Acquired From Indoor Treadwear Test
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
Appendix 4. Raw Dry Traction Testing Results - Asphalt
Barcode Candidate Test Brand Tire Line Tire Size Average STDev High Low Average STDev High Low
Tire Number Slide Slide Slide Slide Peak Peak Peak Peak
Number
POTENZA
3141 080211 0830063 Bridgestone RE92 P225/60R16 59.3 3.0 64.8 56.4 95.3 1.7 97.9 92.9
POTENZA
3498 080113 0830065 Goodyear INTEGRITY P225/60R17 65.1 1.0 66.2 63.4 97.0 1.0 98.5 96.0
PILOT HX
3627 080131 0830074 Michelin MXM4 225/60R16 52.7 1.1 54.1 51.6 99.3 1.1 100.7 98.2
PILOT HX
3633 080132 0830075 Michelin MXM4 225/60R16 54.8 2.1 57.4 52.5 101.1 1.9 103.6 98.2
TIGER PAW
ASTM F 2493
3727 080145 0830076 Uniroyal SRTT P225/60R16 67.0 1.9 68.8 64.7 99.4 3.1 103.8 96.3
TIGER PAW
ASTM F 2493
3733 080146 0830077 Uniroyal SRTT P225/60R16 66.4 2.5 68.7 61.9 99.7 2.0 103.0 97.1
POTENZA
3161 080151 0830079 Bridgestone RE750 225/60R16 55.1 0.6 56.1 54.2 102.7 1.8 105.3 99.9
POTENZA
3167 080156 0830080 Bridgestone RE750 225/60R16 57.5 1.5 60.2 56.0 105.1 4.9 110.6 98.6
P6 FOUR
3702 080161 0830081 Pirelli SEASONS 225/60R16 68.2 3.0 72.0 64.3 104.3 3.2 108.0 100.3
P6 FOUR
3211 080166 0830085 Bridgestone TURANZA LS-V P225/60R16 73.7 3.1 78.4 70.6 101.7 1.4 103.7 99.3
LIFELINER
3325 080171 0830087 Cooper TOURING SLE 225/60R16 62.1 2.2 65.9 60.3 94.2 2.6 98.1 90.3
124
Number
LIFELINER
No brand
3344 080175 0830090 name WINTERFORCE 225/60R16 66.2 1.7 68.7 64.1 90.8 3.1 95.5 87.2
No brand
SP 4000
3369 080182 0830092 Dunlop DSSST C P225/60R17 66.7 2.4 70.5 63.4 91.4 2.3 94.7 88.7
SP 4000
3375 080185 0830093 Dunlop DSSST C P225/60R17 67.7 2.7 71.3 64.9 92.1 1.0 93.2 90.5
BLIZZAK
REV01
3111 080187 0830096 Bridgestone STUDLESS 225/60R16 77.7 2.8 80.2 72.4 94.1 1.2 95.4 92.4
BLIZZAK
REV01
3419 080104 0830098 Goodyear INTEGRITY 225/60R16 66.6 1.1 67.7 64.6 95.6 3.1 101.8 92.2
3683 080203 0830101 Futura TOURING HR P225/60R16 57.8 1.8 60.3 55.3 95.2 2.7 98.2 91.8
3192 080197 0830103 Bridgestone TURANZA LS-T P225/60R16 56.3 1.6 58.4 54.5 95.2 2.4 98.2 92.8
125
Appendix 5. Raw Dry Traction Testing Results - Concrete
Number
BLIZZAK
REV01
BLIZZAK
REV01
POTENZA
POTENZA
POTENZA
3161 080151 0830079 Bridgestone RE750 225/60R16 72.4 3.6 76.8 68.5 105.8 3.8 111.6 101.9
POTENZA
3167 080156 0830080 Bridgestone RE750 225/60R16 71.5 4.9 76.2 64.6 110.6 5.2 117.4 103.2
LIFELINER
LIFELINER
No brand
No brand
SP 4000
3369 080182 0830092 Dunlop DSSST C P225/60R17 79.3 0.6 80.1 78.4 102.2 3.3 107.4 98.7
SP 4000
3375 080185 0830093 Dunlop DSSST C P225/60R17 80.1 0.7 81.3 79.4 98.2 1.8 101.5 96.5
126
Number
PILOT HX
3627 080131 0830074 Michelin MXM4 225/60R16 66.8 4.3 73.6 62.6 106.8 2.1 108.5 102.9
PILOT HX
3633 080132 0830075 Michelin MXM4 225/60R16 72.5 2.8 76.1 67.5 104.5 5.7 112.3 96.8
P6 FOUR
P6 FOUR
TIGER PAW
ASTM F 2493
3727 080145 0830076 Uniroyal SRTT P225/60R16 81.8 2.1 84.9 79.5 106.5 5.4 115.7 102.5
TIGER PAW
ASTM F 2493
3733 080146 0830077 Uniroyal SRTT P225/60R16 81.6 0.8 82.5 80.3 104.5 4.3 111.7 100.4
127
Appendix 6. Raw Wet Traction Testing Results - Asphalt
Number
3626 080133 0830004 Michelin PILOT HX 225/60R16 51.3 1.3 54.1 49.1 94.7 3.8 101.0 89.9
MXM4
MXM4
3726 080144 0830014 Uniroyal TIGER PAW P225/60R16 58.3 0.8 59.5 57.1 94.5 2.7 98.3 91.2
ASTM F 2493
SRTT
ASTM F 2493
SRTT
3166 080152 0830022 Bridgestone POTENZA 225/60R16 58.8 0.7 59.9 57.6 95.9 3.0 102.0 92.4
RE750
RE750
3116 080191 0830028 Bridgestone BLIZZAK 225/60R16 50.0 0.7 51.4 49.2 80.8 4.0 88.6 75.2
REV01
STUDLESS
REV01
STUDLESS
3349 080176 0830032 No brand WINTERFORCE 225/60R16 52.5 0.5 53.2 51.5 78.4 3.4 84.7 73.9
name
name
128
Number
3707 080158 0830040 Pirelli P6 FOUR 225/60R16 59.8 0.7 61.1 58.8 86.9 3.0 91.5 82.7
SEASONS
3701 080159 0830041 Pirelli P6 FOUR 225/60R16 61.1 0.6 62.0 59.9 87.0 2.0 91.1 84.9
SEASONS
3318 080172 0830044 Cooper LIFELINER 225/60R16 54.5 0.7 55.9 53.8 89.9 4.1 96.3 84.0
TOURING SLE
TOURING SLE
3374 080181 0830046 Dunlop SP 4000 P225/60R17 53.5 0.8 54.7 52.2 88.6 2.6 93.5 83.9
DSSST C
3368 080186 0830047 Dunlop SP 4000 P225/60R17 53.8 0.5 54.6 52.5 86.4 2.0 90.0 83.4
DSSST C
3135 080210 0830060 Bridgestone POTENZA P225/60R16 45.7 0.5 46.6 44.9 86.0 4.4 89.8 77.9
RE92
RE92
129
Appendix 7. Raw Wet Traction Testing Results - Concrete
Number
MXM4
MXM4
ASTM F 2493
SRTT
ASTM F 2493
SRTT
RE750
RE750
REV01
STUDLESS
REV01
STUDLESS
name
name
130
Number
3707 080158 0830040 Pirelli P6 FOUR 225/60R16 38.5 0.9 40.1 37.1 64.5 3.6 70.5 58.9
SEASONS
3701 080159 0830041 Pirelli P6 FOUR 225/60R16 39.6 1.0 40.6 37.7 64.6 3.4 69.9 59.6
SEASONS
TOURING SLE
TOURING SLE
3374 080181 0830046 Dunlop SP 4000 P225/60R17 40.2 1.2 41.8 38.2 64.8 2.8 70.1 61.2
DSSST C
3368 080186 0830047 Dunlop SP 4000 P225/60R17 40.1 1.1 41.4 38.0 65.0 2.1 67.8 62.4
DSSST C
RE92
RE92
131
Appendix 8. UTQG Adjusted Wet Traction Testing Results
Barcode Candidate Test Brand Tire Line Tire Size Raw Raw UTQG UTQG Labeled Attained
Tire Number Asphalt Concrete Adjusted Adjusted Traction Traction
Number Average Average Asphalt Concrete Grade Grade
Slide Slide Slide Slide
Coefficient Coefficient Coefficient Coefficient
3626 080133 0830004 Michelin PILOT HX 225/60R16 0.513 0.404 0.49 0.39 A A
MXM4
3632 080137 0830005 Michelin PILOT HX 225/60R16 0.504 0.398 0.48 0.39 A A
MXM4
3726 080144 0830014 Uniroyal TIGER PAW P225/60R16 0.583 0.395 0.55 0.38 A AA
ASTM F 2493
SRTT
3734 080147 0830015 Uniroyal TIGER PAW P225/60R16 0.593 0.396 0.56 0.38 A AA
ASTM F 2493
SRTT
3503 080114 0830016 Goodyear INTEGRITY P225/60R17 0.498 0.366 0.47 0.37 A A
3424 080105 0830020 Goodyear INTEGRITY 225/60R16 0.492 0.354 0.47 0.35 A B
3418 080109 0830021 Goodyear INTEGRITY 225/60R16 0.485 0.348 0.46 0.35 A B
3166 080152 0830022 Bridgestone POTENZA 225/60R16 0.588 0.420 0.55 0.41 AA AA
RE750
3160 080155 0830023 Bridgestone POTENZA 225/60R16 0.594 0.425 0.56 0.42 AA AA
RE750
3116 080191 0830028 Bridgestone BLIZZAK 225/60R16 0.500 0.374 0.46 0.39 none B
REV01
STUDLESS
3110 080192 0830029 Bridgestone BLIZZAK 225/60R16 0.490 0.373 0.45 0.36 none B
REV01
STUDLESS
3349 080176 0830032 No brand WINTERFORCE 225/60R16 0.525 0.355 0.49 0.35 none B
name
3343 080180 0830033 No brand WINTERFORCE 225/60R16 0.522 0.352 0.48 0.34 none B
name
3707 080158 0830040 Pirelli P6 FOUR 225/60R16 0.598 0.385 0.55 0.37 A A
SEASONS
132
Barcode Candidate Test Brand Tire Line Tire Size Raw Raw UTQG UTQG Labeled Attained
Tire Number Asphalt Concrete Adjusted Adjusted Traction Traction
Number Average Average Asphalt Concrete Grade Grade
Slide Slide Slide Slide
Coefficient Coefficient Coefficient Coefficient
3701 080159 0830041 Pirelli P6 FOUR 225/60R16 0.611 0.396 0.57 0.38 A AA
SEASONS
3318 080172 0830044 Cooper LIFELINER 225/60R16 0.545 0.402 0.50 0.39 A A
TOURING SLE
3320 080174 0830045 Cooper LIFELINER 225/60R16 0.545 0.388 0.50 0.37 A A
TOURING SLE
3374 080181 0830046 Dunlop SP 4000 P225/60R17 0.535 0.402 0.50 0.38 A A
DSSST C
3368 080186 0830047 Dunlop SP 4000 P225/60R17 0.538 0.401 0.50 0.38 A A
DSSST C
3185 080194 0830050 Bridgestone TURANZA LS-T P225/60R16 0.568 0.408 0.53 0.39 A A
3191 080200 0830051 Bridgestone TURANZA LS-T P225/60R16 0.586 0.411 0.55 0.39 A AA
3218 080163 0830052 Bridgestone TURANZA LS-V P225/60R16 0.595 0.429 0.57 0.41 AA AA
3212 080165 0830053 Bridgestone TURANZA LS-V P225/60R16 0.582 0.415 0.55 0.40 AA AA
3676 080206 0830056 Futura TOURING HR P225/60R16 0.547 0.414 0.53 0.40 A A
3682 080208 0830057 Futura TOURING HR P225/60R16 0.538 0.405 0.52 0.39 A A
3135 080210 0830060 Bridgestone POTENZA P225/60R16 0.457 0.360 0.44 0.34 A B
RE92
3142 080214 0830061 Bridgestone POTENZA P225/60R16 0.471 0.367 0.46 0.35 A B
RE92
133
Appendix 9. ASTM E501 Reference Tire Wet Traction Testing Results
Asphalt Concrete
Average Slide Average Peak Average Slide Average Peak
Tire Test Tire
Number Number Tire Line Size Slide STDev Peak STDev Slide STDev Peak STDev Candidate Test Numbers
ASTM
080022 0831006 G78-15 51.8 0.6 91.4 3.7 36.3 0.7 56.6 4.4
E501
0830004 0830005 0830008 0830009
ASTM
080021 0831007 G78-15 52.8 0.6 90.0 3.9 36.1 1.0 55.0 5.9
E501
ASTM
080021 0831012 G78-15 54.3 0.2 91.9 3.4 36.4 1.0 56.4 4.0
E501
0830010 0830011 0830014 0830015
ASTM
080022 0831013 G78-15 53.3 0.5 90.4 4.5 36.0 0.8 57.2 7.6
E501
ASTM
080022 0831018 G78-15 52.6 0.4 87.0 3.6 35.3 0.8 57.4 5.6
E501
0830016 0830017 0830020 0830021
ASTM
080021 0831019 G78-15 52.4 0.4 86.1 4.0 34.7 0.7 57.1 5.5
E501
ASTM
080021 0831024 G78-15 54.2 0.3 81.4 1.4 35.6 0.6 57.7 4.5
E501
0830022 0830023 0830026 0830027
ASTM
080022 0831025 G78-15 53.5 0.6 81.5 1.7 35.7 0.7 57.0 3.3
E501
ASTM
080022 0831030 G78-15 54.3 0.4 84.8 2.9 36.2 0.6 54.3 2.0
E501
0830028 0830029 0830032 0830033
ASTM
080021 0831031 G78-15 53.6 0.6 83.7 3.1 35.7 0.5 53.1 2.5
E501
ASTM
080021 0831036 G78-15 54.8 0.6 80.9 4.9 35.6 0.6 52.0 3.7
E501
0830034 0830035
ASTM
080022 0831037 G78-15 54.6 0.6 78.0 2.6 35.6 0.5 54.4 2.2
E501
ASTM
080022 0831042 G78-15 54.4 0.6 83.4 3.9 36.6 0.6 51.7 3.6
E501
0830040 0830041 0830044 0830045
ASTM
080021 0831043 G78-15 54.3 0.4 85.3 3.2 36.2 0.4 60.2 4.2
E501
ASTM
080021 0831048 G78-15 54.2 0.6 87.4 4.6 37.4 1.0 59.6 3.8
E501
0830046 0830047 0830050 0830051
ASTM
080022 0831049 G78-15 52.7 0.4 88.4 2.3 36.5 0.7 56.7 4.9
E501
ASTM
080022 0831054 G78-15 53.0 0.5 87.2 3.6 36.7 0.5 59.6 3.9
E501
0830052 0830053
ASTM
080021 0831055 G78-15 53.0 0.5 87.2 3.6 36.7 0.5 59.6 3.9
E501
ASTM
080021 0831058 G78-15 52.1 0.4 84.6 3.4 36.8 0.4 57.1 5.5
E501
0830056 0830057 0830060 0830061
ASTM
080022 0831059 G78-15 51.0 0.3 84.7 3.7 36.4 0.8 56.4 6.3
E501
134
Appendix 10. ASTM E501 Reference Tire Dry Traction Testing Results
Asphalt Concrete
Tire Test Average Slide Average Peak Average Slide Average Peak
Number Number Tire Line Tire Size Slide STDev Peak STDev Slide STDev Peak STDev
080081 0831062 ASTM E501 G78-15 67.5 1.4 69.9 66.1 81.8 0.8 83.2 80.6
080081 0831078 ASTM E501 G78-15 62.7 6.2 70.7 56.3 81.2 1.4 82.8 79.6
080081 0831094 ASTM E501 G78-15 61.7 2.2 66.1 59.9 76.3 3.7 81.7 70.9
080082 0831067 ASTM E501 G78-15 68.3 2.5 72.4 65.2 82.0 1.1 83.3 80.5
080082 0831083 ASTM E501 G78-15 64.0 2.9 68.2 59.1 81.5 0.6 82.3 80.8
080082 0831095 ASTM E501 G78-15 62.1 3.3 65.9 57.2 82.0 1.7 84.7 79.8
080083 0831072 ASTM E501 G78-15 66.8 2.4 69.6 64.2 80.4 1.2 82.1 78.7
080083 0831084 ASTM E501 G78-15 59.0 1.8 61.9 56.7 81.1 2.8 84.9 78.3
080083 0831100 ASTM E501 G78-15 59.2 0.8 60.7 58.6 78.1 3.0 83.2 74.3
080084 0831073 ASTM E501 G78-15 67.1 1.8 68.9 64.4 82.8 1.5 84.9 81.1
080084 0831089 ASTM E501 G78-15 63.1 4.0 67.2 57.9 79.0 2.0 81.2 76.5
080084 0831105 ASTM E501 G78-15 60.4 1.5 62.5 58.9 77.5 4.2 81.9 71.8
135
REFERENCES
1. Hall, D.E., & Moreland, J.C. (2001). Fundamentals of Rolling Resistance. Rubb. Chem.
Technol., 74 No 3, 525.
2. Rubber Manufacturers Association. (2008). Statistics: North American Marketplace Tire
Shipments, U.S. and Canada, Updated February 2008. Washington, DC: Rubber
Manufacturers Association. Available at www.TireBusiness.com.
3. United States Congress. (2003). Making Appropriations for Agriculture, Rural Development,
Food and Drug Administration, and Related Agencies for the Fiscal Year Ending
September 30, 2004, and for Other Purposes. November 25, 2003, p. 971. United
Conference Report 108-401, to Accompany H.R. 2673. Washington, DC: United States
Congress.
4. National Research Council of the National Academies (2006). Tires and Passenger Vehicle
Fuel Economy. Special Report 286, Washington, DC: Transportation Research Board.
5. LaClair, T. J., Rolling Resistance, p. 476-477, in The Pneumatic Tire, Gent, A.N., & Walter,
J.D. (Ed.). (2006). DOT HS 810 561. Published under contract DTNH22-02-P-07210.
Washington, DC: National Highway Traffic Safety Administration.
6. LaClair, p. 477.
7. Department of Energy. (2009). Advanced Technologies and Energy Efficiency; Where Does
the Energy Go? Washington, DC: Department of Energy. Available at Web site:
http://www.fueleconomy.gov/feg/atv.shtml.
8. Duleep, K.G. (2005). Tires, Technology, and Energy Consumption. PowerPoint Presentation.
Energy Efficient Tyres: Improving the On-Road Performance of Motor Vehicles, 15-16
November, 2005, International Energy Agency, Paris, France.
9. National Research Council, p. 30. Washington, DC: Transportation Research Board.
10. Reimpell, J., Stoll, H., & Betzler, J. W. (2001). Published in: The Automotive Chassis:
Engineering Principles, 2nd Edition, p. 123. Reprinted by SAE International, Warrendale,
PA. Oxford, UK: Reed Elsevier PLC Group .
11. LaClair, p. 501.
12. National Research Council, p. 33.
13. NHTSA. (2006). Traffic Safety Facts 2005. A Compilation of Motor Vehicle Crash Data
from the Fatality Analysis Reporting System and the General Estimates System. DOT HS
810 631. Washington, DC: National Highway Traffic Safety Administration. http://www-
nrd.nhtsa.dot.gov/Pubs/TSF2005.PDF.
15. United States Congress. (2007). Energy Independence and Security Act of 2007. Section 111,
Consumer Tire Information.
16. Evans, L. R., & W. H. Waddell. (1995). Kautsch. Gummi Kunstst., Vol. 48, p. 718.
17. Wolff, S. (1982). Rubb. Chem. Technol., Vol. 55, p. 967.
18. Futamura, S. (1990). Rubb. Chem. Technol., Vol. 64, p. 57.
19. Futamura, S. (1990) Tire Science and Technology, TSTCA, Vol. 18, p. 2.
20. CFR Title 40, §86; Title 40--Protection of Environment, Chapter I--Environmental Protection
Agency (Continued), Part 86--Control Of Emissions From New And In-Use Highway
Vehicles And Engines. Washington, DC: Code of Federal Regulations.
http://www.access.gpo.gov/nara/cfr/waisidx_07/40cfr86_07.html.
136
21. EPA (2009). Many Factors Affect MPG. Washington, DC: Environmental Protection
Agency. Available at http://www.fueleconomy.gov/feg/factors.shtml.
22. EPA (2009). Detailed Test Information. Washington, DC: Environmental Protection Agency.
Available at http://www.fueleconomy.gov/feg/fe_test_schedules.shtml.
23. CFR § 86.162–00 Approval of alternative air conditioning test simulations and descriptions
of AC1 and AC2. Washington, DC: Code of Federal Regulations.
http://edocket.access.gpo.gov/cfr_2008/julqtr/pdf/40cfr86.162-00.pdf.
24. NHTSA, (2009). Tire Ratings. Washington, DC: National Highway Traffic Safety
Administration. Available at http://www.safercar.gov.
25. NHTSA, Tire Ratings.
26. Grosch, K. A., & Schallamach, A. (1961). Tyre Wear at Controlled Slip. Wear, 4, pp. 356 –
371.
27. National Research Council.
28. NHTSA (2001). Tire Pressure Special Study - Vehicle Observation Data, August, 2001. DOT
HS 809 317. Washington, DC: National Highway Traffic Safety Administration.
29. NHTSA (2009). Tire Pressure Maintenance – A Statistical Investigation, April, 2009. DOT
HS 811 086. Washington, DC: National Highway Traffic Safety Administration.
30. Wicks, F. & Sheets, W. (1991). “Effect of Tire Pressure and Performance Upon Oil Use and
Energy Policy Options.” Proceedings of the 26th Intersociety Energy Conversion
Engineering Conference IECEC-91, August 4-9, 1991, Boston, Massachusetts. Volume
4, pp. 307. La Grange, IL: American Nuclear Society.
31. EPA. (2009). Fuel Economy Guide. Washington, DC: Environmental Protection Agency.
[Website]. http://www.fueleconomy.gov/feg/maintain.shtml.
32. Clark, S. K., & Dodge, R. N. (1979). A Handbook for the Rolling Resistance of Pneumatic
Tires. Prepared for the U.S. Department of Transportation. Ann Arbor: Regents of the
University of Michigan.
33. Hall, D. E., & Moreland, J. C. (2000). Fundamentals of Rolling Resistance. Spring 2000
Education Symposium No. 47, Basic Tire Technology: Passenger and Light Truck.
Akron, OH: American Chemical Society.
34. Continental Tire. (2008, November 17). Government Regulation in Transition – Continental
Tire Point of View. Presentation before the California Energy Commission. Available at
http://www.energy.ca.gov/transportation/tire_efficiency/documents/2008-11
17_roundtable/presentations/Continental%20Tire%20Presentation.pdf.
35. Grugett, B. C., Martin, E. R., & Thompson, G. D. (1981). The Effects of Tire Inflation
Pressure on Passenger Car Fuel Consumption. International Congress and Exposition,
February 23-27, 1981. Paper # 810069, SAE Technical Paper Series. Warrendale, PA:
Society of Automotive Engineers, Inc.
39. Luchini, J. R. (1983). Rolling Resistance Test Methods. In Tire Rolling Resistance, Rubber
Division Symposia Vol. 1, D. J. Schuring, Ed. Akron, OH: American Chemical Society
40. ISO. (2009). DIS ISO 28580 Draft International Standard – Tyre Rolling Resistance
measurement method – single-point test and measurement result correlation – designed to
137
facilitate international cooperation and, possibly, regulation building. Passenger Car,

Truck and Bus Tyres. Geneva, Switzerland: International Organization for
Standardization.
41. Schuring, D. J., & Futamura, S. (1990). Rolling Loss of Pneumatic Highway Tires in the
Eighties. Rubber Chemistry and Technology, Vol. 62, No. 3, pp. 315–367.
43. ISO. (2005). ISO 18164, T Passenger car, truck, bus and motorcycle tyres — Methods of
measuring rolling resistance. Geneva, Switzerland: International Organization for
Standardization.
44. Clark & Dodge, p.7.
45. Clark & Dodge, p.7.
46. SAE (2006). J1269 - Surface Vehicle Recommended Practice for Rolling Resistance
Measurement Procedure for Passenger Car, Light Truck, and Highway Truck and Bus
Tires. Issued 1979-11, Revised 2006-09, Superseding J1269 SEP2000, p. 5. Warrendale,
PA: Society of Automotive Engineers.
47. SAE (2006). J2452 - Surface Vehicle Recommended Practice for Stepwise Coastdown
Methodology for Measuring Tire Rolling Resistance. Issued 1999-06. p. 8. Warrendale,
PA: Society of Automotive Engineers.
48. Lambillotte, B. (2009, February 5). California Energy Commission’s Fuel Efficient Tire
Program. PowerPoint Presentation. Akron, OH: Smithers Scientific Services, Inc.
49. Schuring & Futamura, pp. 315–367.
50. The Tire Rack (2009). Tire Tech Information/General Tire Information. P-Metric and Euro
Metric Tire Sizing. http://www.tirerack.com/tires/tiretech/techpage.jsp?techid=24.
51. Rubber Manufacturers Association. (2009). Comments TN-48720.pdf, to the April 8, 2009
California Energy Commission Staff Workshop on the Fuel Efficient Tire Program,
http://www.energy.ca.gov/transportation/tire_efficiency/documents/2009-04
08_workshop/comments/.
53. Pillai, P.S. (1995). Total Tire Energy Loss Comparison by the Whole Tire Hysteresis and the
Rolling Resistance Methods. Tire Science and Technology, TSTCA, Vol. 23, No. 4, pp.
256-265.
54. The Tire & Rim Association. (2004). Engineering Design Information for Ground Vehicle
Tires, Pages 1-11 & 1-15, Rev. 5. http://www.us-tra.org/traPubs.html.
138
DOT HS 811 154

August 2009

Tyre Paper PDF

Uploaded by

Copyright:

Available Formats

Tyre Paper PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tyre Paper PDF

Uploaded by

Copyright:

Available Formats

What is the purpose of this document?

What is the purpose of this document?

What topics are covered in the document?

What topics are covered in the document?

DOT HS 811 154 August 2009

NHTSA Tire Fuel Efficiency

1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.

P.O. Box B-37 DTNH22-03-D-08660,

Washington, DC 20590 NHTSA/NVS-312

in inches 2.54 centimeters cm mm millimeters 0.04 inches in

MASS (weight) MASS (weight)

oz ounces 28 grams g g grams 0.035 ounces oz

TEMPERATURE (exact) TEMPERATURE (exact)

1.0 INTRODUCTION ................................................................................................................................ 1

1.1 THE CONCEPT OF ROLLING RESISTANCE .................................................................................... 3

2.0 METHODOLOGY ............................................................................................................................... 7

2.1 TEST TIRES .................................................................................................................................. 7

2.1.1 ASTM F2493 Radial Standard Reference Test Tire ............................................................... 7

2.2 TIRE ROLLING RESISTANCE TEST PROCEDURES ......................................................................... 8

2.2.2 SAE J1269 & ISO 18164 Multi-Point Rolling Resistance.................................................... 11

2.2.3 SAE J2452 Multi-Point (Speed Coast Down) Rolling Resistance........................................ 11

2.3 FUEL ECONOMY TEST VEHICLE ................................................................................................ 11

2.4 TEST WHEELS ............................................................................................................................ 11

2.5 TEST MATRIX ............................................................................................................................ 12

2.6 TREAD COMPOUND PROPERTIES TESTING................................................................................. 13

2.7 ON-VEHICLE FUEL ECONOMY TESTING .................................................................................... 16

2.7.1 EPA 40 CFR Part 86 Dynamometer Fuel Economy Testing ................................................ 17

2.8 SKID-TRAILER TIRE TRACTION TESTING .................................................................................. 21

2.9 ON-VEHICLE TIRE TREADWEAR TESTING ................................................................................. 23

2.10 INDOOR TIRE TREADWEAR TESTING ......................................................................................... 25

3.1 EFFECT OF TIRE ROLLING RESISTANCE ON AUTOMOBILE FUEL EFFICIENCY ........................... 28

3.1.1 Preliminary Analysis: Data Shifts ......................................................................................... 30

3.1.2 Highway FET Triplicate Analysis: ....................................................................................... 31

3.1.3 Air Conditioning SC03 – 11/20/08 to 11/25/08 .................................................................... 34

3.1.5 Effect of Tire Rolling Resistance on Fuel Economy............................................................. 37

3.1.6 Effect of Reduced Inflation Pressure on Fuel Economy ....................................................... 43

3.1.7 Fuel Economy Testing Summary.......................................................................................... 50

3.2 CORRELATION OF TANGENT Δ AT 60°C TO TIRE ROLLING RESISTANCE .................................. 51

3.3 EFFECT OF TIRE ROLLING RESISTANCE ON SAFETY.................................................................. 53

3.3.1 Dry Traction Data ................................................................................................................. 53

3.3.2 Wet Traction Data ................................................................................................................. 56

3.3.3 UTQGS Traction Grade ........................................................................................................ 59

3.3.4 Correlation of Tangent δ at 0°C to Wet Traction Properties................................................. 61

3.4 EFFECTS OF TIRE ROLLING RESISTANCE ON TREADWEAR RATE .............................................. 62

3.4.1 Analysis of Wear Data From Indoor Treadwear Testing ...................................................... 65

5.0 REQUIREMENTS ............................................................................................................................. 79

6.0 ROLLING RESISTANCE (Fr) VERSUS ROLLING RESISTANCE COEFICIENT Cr)......... 85

6.1 THEORY OF FR AND CR .............................................................................................................. 85

6.2 DISCUSSION ............................................................................................................................... 94

Appendix 2. Detailed Test Matrix ..................................................................................................... 101

Appendix 4. Raw Dry Traction Testing Results - Asphalt.............................................................. 124

Appendix 5. Raw Dry Traction Testing Results - Concrete ........................................................... 126

Appendix 6. Raw Wet Traction Testing Results - Asphalt ............................................................. 128

Appendix 7. Raw Wet Traction Testing Results - Concrete........................................................... 130

Appendix 8. UTQG Adjusted Wet Traction Testing Results ......................................................... 132

Figure 1. Where Does the Energy Go? ........................................................................................... 4

Figure 3. Magic Triangle: Traction, Treadwear, and Rolling Resistance....................................... 6

Figure 4. Force Method Rolling Resistance Test Machine........................................................... 10

Figure 5. Torque Method Rolling Resistance Test Machine ........................................................ 10

SAE – The Society of Automotive Engineers International is an international standards organiza

Fr = Calculated Rolling Resis