Académique Documents
Professionnel Documents
Culture Documents
SAE TECHNICAL
PAPER SERIES 2006-01-0559
Gary J. Heydinger
The Ohio State University
Paul A. Grygier
National Highway Traffic Safety Administration
400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760 Web: www.sae.org
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
SAE Permissions
400 Commonwealth Drive
Warrendale, PA 15096-0001-USA
Email: permissions@sae.org
Tel : 724-772-4028
Fax : 724-776-3036
ISSN 0148-7191
Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely
responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in
SAE Transactions.
Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300
word abstract to Secretary, Engineering Meetings Board, SAE.
Printed in USA
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
2006-01-0559
Gary J. Heydinger
The Ohio State University
Paul A. Grygier
National Highway Traffic Safety Administration
Abstract The data set from these tests was used to create a tire model
for a dynamic simulation. The paper concludes by displaying
There exists a fairly extensive set of tire force measurements the model’s longitudinal and lateral forces versus loading
performed on dry pavement. But in order to develop a low- condition and tire slip angle.
coefficient of friction tire model, a set of tire force
measurements made on wet pavement is required. Using Introduction
formulations and parameters obtained on dry roads, and then
reducing friction level to that of a wet road is not sufficient to The purpose of this research was to develop a low coefficient
model tire forces in a high fidelity simulation. This paper of friction tire model. The model was used on the National
describes the process of more accurately modeling low Advanced Driving Simulator (NADS) for a study to
coefficient tire forces on the National Advanced Driving investigate the safety benefits of Electronic Stability and
Simulator (NADS). It is believed that the tire model Control (ESC) systems. The research was not to develop ESC
improvements will be useful in many types of NADS systems, but rather to use an existing ESC system to study
simulations, including ESC and other advanced vehicle drivers’ performances from a human factors perspective. The
technology studies. low coefficient of friction tire model was needed to increase
the incidences of ESC activation as test subjects drove
In order to produce results that would come from a road through the various NADS scenarios. The vehicle modeled
surface that would be sufficiently slippery, a set of tires were was a 2002 Oldsmobile Intrigue with Goodyear Eagle RSA
shaved to 4/32 inches and sent to a tire-testing lab for P225/60R16 tires. To make the NADS study realistic and
measurement. Shaving a tire does not produce the same useful, the vehicle dynamics model must be of high fidelity;
effects that would come about by allowing a tire to wear down that is, the physics predicted through the simulation should be
to 4/32 inches through normal use. However, for this study, very close to real-world experiences.
only the coefficient of friction needed to be reduced. The
aging effects of rubber are ignored. Tire forces were NADS vehicle dynamics have been validated with various
measured on a tire test machine, using a water-coated surface vehicles (1994 Ford Taurus, 1998 Chevrolet Malibu, 1997
to approximate the frictional properties of wet pavement. Jeep Cherokee, and 1991 GM-Volvo heavy truck with 1992
These tests, which included cornering, braking, and driving, Fruehauf trailer), but not with cases involving low coefficients
were performed at five loading conditions. They showed a of friction, like driving on wet roads and ice. In order to
decrease in tire effective lateral stiffness as well as a drop in properly model the low coefficient surface, tire tests were
longitudinal force as the tire speed increased. performed under low friction conditions. Modeling tire
mechanics for vehicle dynamics relies heavily on tire testing,
and most models are dominated by empirical formulations.
1
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
Using formulations and parameters obtained on dry roads, and shows peak longitudinal coefficient of friction data from the
then simply scaling the friction level and associated forces and 4/32” tread depth tire. The curves in Figure 2 bound our
moments to that of a lower coefficient road, is not sufficient to target peak friction value of about 0.5 at 50 mph. We selected
accurately replicate tire forces, particularly when the vehicle is a water depth of 0.05” for our tests, with the expectation that
operating at highway speeds. Therefore, this research we would get similar peak longitudinal force values from our
involved measuring tire forces at different speeds on a low tests, suiting the needs of our driving simulator research. A
friction, wet test surface, and to developing a tire model with tread depth of 4/32” represents a moderately well worn tire.
speed dependency. The tire model based on these NHTSA studies have indicated that the average tread depth
measurements provides realistic forces in the linear and for in-service tires, based on measurements made on 11,530
nonlinear range, and the peak friction is at a level where ESC vehicles, to be 7/32” [2]. Shaving a tire does not produce the
engages during driving tasks specified by the NADS ESC same effects that would come about by allowing a tire to wear
study testing protocol. down to 4/32 inches through normal use. However, for this
study, only the coefficient of friction needed to be reduced.
For the NADS ESC study, the goal was to have the peak tire
coefficient of friction be less than 0.6 to ensure ESC From SAE Paper 2002-01-0553
1.2
activation. This condition can be produced in the “tire-
laboratory” with a shaved tire running on a wet test surface.
Variations of tread depth, water depth, tire pressure, tire
1
construction, surface texture, and tread patterns were not
model.
The tire model developed in this research was used with the 1
3
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
The loss of shear force capabilities at high speed on flooded Speed Increases
and the tire contact area. Tread pattern and surface macro and
Peak (Fy/Fz)
micro texture are very important in wet surface contact. The 30 mph Wet
particular when tire forces are close to saturation, or in sliding 75 mph Wet
0
mode. 0 500 1000 1500 2000 2500
Fz (lbs)
4
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
1.5
Speed Increases
30 mph Dry -4
x 10
30 mph Dry
3
30 mph Wet
0 Mx=(KLT0+KLT1*Fz+KLT2*Fz 2)*Fy
300 400 500 600 700 800 900 1000 1100 1200
0.5 Fz (lbs)
-4
x 10
60 mph Wet
Figure 6. Longitudinal Peak Coefficient of Friction 400 600 800 1000 1200 1400 1600 1800 2000 2200
Fz (lbs)
The effective tire lateral stiffness at different speeds is shown Figure 9. Effective Overturning Moment Stiffness
in Figure 7, the effective aligning moment stiffness in Figure
8, and the effective overturning moment stiffness in Figure 9.
Figures 10-11 show the lateral force and overturning moment
35
inclination angle stiffnesses at all test speeds and vertical
loads. 30 Speed Increases
25
400
20
CJ (lbs/deg)
350
30 mph Dry
15
300
10
Cornering Stiffness (lbs/deg)
250 5
30 mph Dry
0 30 mph Wet
200 30 mph Wet 45 mph Wet
-5 60 mph Wet
150 75 mph Wet
30 mph Dry -10
30 mph Wet 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
100 45 mph Wet
45 mph Wet Fz (lbs)
60 mph Wet
50 60 mph Wet 75 mph Wet Figure 10. Lateral Force-Inclination Angle Stiffness
75 mph Wet
0
0 500 1000 1500 2000 2500
Fz (lbs) 0.01
-1
0.005
-1.2 0.004
0.003
-1.4
KK1 (ft/lb)
0.002
Mz=KK1*Fy*Fz
-1.6
0.001
30 mph Dry
30 mph Wet 0
-1.8 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
45 mph Wet
60 mph Wet Fz (lbs)
-2 75 mph Wet
Figure 11. Overturning Moment-Inclination Angle Stiffness
-2.2
400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 The experimental tire data is fitted with the empirical STI
Fz (lb)
saturation function given below, and the results for dry and
Figure 8. Effective Aligning Moment Stiffness wet conditions at 30 mph are shown in Figures 12 and 13.
5
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
2000
1.4 1500
Test
1000
Fitted
1.2
1 0
Fz = 457.9 lbs
-500
0.8
Fz = 919.8 lbs
-1000
f(V )
Fz = 1380 lbs
0.6 -1500
Fz = 1842 lbs
-2000
0.4 Fz = 2306 lbs
-2500
-25 -20 -15 -10 -5 0 5 10 15 20 25
0.2 Slip Angle (deg)
0
Figure 14. Lateral Forces at 30 mph Dry
0 2 4 6 8 10 12 14 16
V
30 MPH WET
Figure 12. Saturation Function For Dry Tests (30 mph) 2500
2000
1.4 1500
Test
1000
Fitted
1.2
Lateral Force (lbs)
500
1 0
Fz = 456.1 lbs
-500
0.8 Fz = 919.3 lbs
-1000
f(V )
Fz = 1380 lbs
0.6 -1500 Fz = 1844 lbs
0
Figure 15. Lateral Forces at 30 mph Wet
0 2 4 6 8 10 12
V
45 MPH WET
2000
Figure 13. Saturation Function For Wet Tests (30 mph) 1500
Wet peak and slide braking and cornering traction coefficients 1000
500
manner as reported by Veith [6]. Our testing confirms this
observation, and adds another dimension: that is; tire normal 0
load variations. Polynomial fits were done for vertical load Fz = 453.9 lbs
variations, and linear interpolation is used to account for speed -500 Fz = 917.3 lbs
-2000
Figures 14-18 show the measured and modeled lateral forces -25 -20 -15 -10 -5 0 5 10 15 20 25
Slip Angle (deg)
versus slip angle for the dry test (30 mph) and four wet tests
done at different speeds (30, 45, 60, and 75 mph). (The 200% Figure 16. Lateral Forces at 45 mph Wet
rated load data was not used for the 60 mph wet test.)
Overall, the model does a good job of predicting the peak
6
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
60 MPH WET (despite some difficulties with the test procedure in this
600
mode). One of the goals for the testing program was to find
400
the combination of tread depth and water depth which would
yield a high coefficient of friction at low speeds and a low
200
coefficient at high speeds. This goal was accomplished. It
should be noted that shaving a tire does not produce the same
Lateral Force (lbs)
Fz = 454.4 lbs
0 effects that would come about by allowing a tire to wear down
Fz = 917.3 lbs to 4/32 inches through normal use. The aging effects of
-200 rubber were ignored.
Fz = 1376 lbs
-400
Fz = 1838 lbs
The tire forces and moments were then used to generate the
tire parameters required by the STI tire model used by the
-600
NADS vehicle simulation dynamics. Some of these
parameters include coefficients for equations describing the
-800
-25 -20 -15 -10 -5 0 5 10 15 20 25 lateral and longitudinal peak coefficient of friction for varying
Slip Angle (deg)
loads at different speeds, effective lateral stiffness, effective
Figure 17. Lateral Forces at 60 mph Wet aligning moment stiffness, several other stiffnesses, and the
tire saturation function.
75 MPH WET
1000
The STI model was then exercised with the generated
800 coefficients. Overall, the model did a good job of predicting
600 peak forces and linear range stiffnesses for all conditions, and
a decent job of modeling the forces when the tires reach high
400
slip angles.
Lateral Force (lbs)
200
Fz = 454.4 lbs
0
Fz = 917.3 lbs
-200
Fz = 1376 lbs References
-400
Fz = 1838 lbs
-600 1. Blythe, W. and Day, T.D., “Single Vehicle Wet Road
Fz = 2303 lbs
-800 Loss of Control; Effects of Tire Tread Depth and
Placement,” SAE Paper 2002-01-0553, 2002.
-1000
-25 -20 -15 -10 -5 0 5 10 15 20 25
Slip Angle (deg)
2. U.S. Dept. of Transportation, National Highway Traffic
Figure 18. Lateral Forces at 75 mph Wet Safety Administration, National Center for Statistical
Analysis, Tire Pressure Special Study - Tread Depth
Figure 19 shows measured and modeled results for Analysis, by K. Thiriez and R. Subramanian, Research
longitudinal, lateral, and combined forces at 30 mph for both Note DOT HS 809 359, Washington, D.C., www-
wet and dry conditions. The model does a good job of nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/RNotes/2001/809-
assimilating the measured results. At 30 mph the wet tire 359TreadDepth.pdf, 2001.
peak longitudinal and lateral forces are only slightly less than
the dry peak forces. 3. Salaani, M.K., “Powertrain and Brake Modeling of the
1994 Ford Taurus for the National Advanced Driving
At higher speeds, the measured results from the combined Simulator,” SAE 980221, 1998.
tests are not consistent due the previously mentioned lack of
quality in the higher speed longitudinal data measurements. 4. Allen, R.W., Rosenthal, T.J., and Chrstos, J.P.,”A Vehicle
Nonetheless, the model predictions for higher speed combined Dynamics Tire Model for Both Pavement and Off-Road
conditions based on peak lateral and longitudinal forces are Conditions,” SAE 970559, 1997.
reasonable.
5. Mechanics of Pneumatic Tires, edited by S.K. Clark,
Conclusions Monograph 122, National Bureau of Standards, 1971.
This paper presented the results from testing shaved passenger 6. Veith, A.G., “Tires – Roads –Rainfall – Vehicles: The
car tires on a low-coefficient test surface. Straight-line and Traction Connection,” Frictional Interaction of Tire and
cornering tests were performed, yielding data for both Pavement, ASTM STP 793. Edited by W. E. Meyer and
longitudinal and lateral forces and cornering moments. Five J. D. Walter, American Society for Testing and Materials,
loading conditions were used, yielding a good fit in the lateral 1983, pp. 3-40.
direction and a reasonable fit in the longitudinal direction
7
Downloaded from SAE International by Steven Sullivan, Wednesday, November 28, 2018
Lateral Slip Angle = +- 2 degrees @ 30 mph Dry Lateral Slip Angle = +- 2 degrees @ 30 mph Wet
1000 Test 1000 Test
Fx (lbs)
Fx (lbs)
0 Model 0 Model
Fz = 1147 lbs
-1000 Fz = 1147 lbs -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
Fy (lbs)
0 0
-1000 -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
Fy (lbs)
0 0
-1000 -1000
-1000 -500 0 500 1000 -1000 -500 0 500 1000
Fx (lbs) Fx (lbs)
Lateral Slip Angle = +- 4 degrees @ 30 mph Dry Lateral Slip Angle = +- 4 degrees @ 30 mph Wet
1000 Test 1000 Test
Fx (lbs)
Fx (lbs)
0 Model 0 Model
Fz = 1147 lbs
Fz = 1147 lbs
-1000 -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
0 Fy (lbs) 0
-1000 -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
Fy (lbs)
0 0
-1000 -1000
-1000 -500 0 500 1000 -1000 -500 0 500 1000
Fx (lbs) Fx (lbs)
Lateral Slip Angle = +- 6 degrees @ 30 mph Dry Lateral Slip Angle = +- 6 degrees @ 30 mph Wet
1000 Test 1000 Test
Fx (lbs)
Fx (lbs)
0 Model 0 Model
Fz = 1147 lbs
Fz = 1155 lbs
-1000 -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
Fy (lbs)
0 0
-1000 -1000
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
Slip Ratio Slip Ratio
1000 1000
Fy (lbs)
Fy (lbs)
0 0
-1000 -1000
-1000 -500 0 500 1000 -1000 -500 0 500 1000
Fx (lbs) Fx (lbs)
Figure 19. Combined Longitudinal and Lateral Forces at 30 mph – Dry and Wet