Standard Tire Model Testing
Standard Tire Model Testing
Standard Tire Model Testing
B a k k e r 4
REFERENCE: van Oosten, J. J. M., Unrau, H.-J., Riedel, A., and Bakker, A., "Stan-
dardization in Tire Modeling and Tire Testing - - TYDEX Workgroup, TIME Project,"
Tire Science and Technology, TSTCA, Vol. 27, No. 3, July-September 1999, pp. 188-
202.
ABSTRACT: As a result of the 1st International Colloquium on Tire Models for Vehicle
Dynamics Analysis in 1991, the international TYDEX Workshop working group was es-
tablished. This workshop concentrated on the standardization of the exchange of tire mea-
surement data and the interface between tire and vehicle models in order to improve the
communications between vehicle manufacturers, suppliers, and research organizations. The
development and knowledge of tire behavior is of great importance to both the tire and
vehicle industries and will be intensified. Therefore the TYDEX Workshop received great
interest from all parties to come to some kind of standardization.
In the two expert groups, one of which focused on Tire Measurements - Tire Modeling
and the other on Tire Modeling - Vehicle Modeling, the TYDEX-Format and the standard
tire interface have been developed, which will be explained in this paper.
Furthermore, a short overview of the European TIME project aiming at a standard tire
testing procedure will be given, which is reliable and consistent with realistic driving
conditions. Standard testing procedures are some of the important consequences of the
TYDEX Workshop.
TNO Road-Vehicles Research Institute, P.O. Box 6033, 2600 JA Delft, The Netherlands.
2 University of Karlsruhe, Institut ftir Mkl and Kfz. Ban, Kaiserstrasse 12, D-76128 Karlsruhe, Germany.
3 IPG, P.O. Box 210522, D-76155 Karlsruhe, Germany.
4 NedCar, P.O. Box 1015, 5700 MC Helmond, The Netherlands.
5 Presented at the seventeenth annual conference of The Tire Society, Akron, Ohio, April 28-29, 1998.
188
van OOSTEN ET AL. ON MODELING, TESTING STANDARDS 189
TABLE 1 - - Possible driving maneuvers and their requirements on the tire modeU
Bandwidth in Hz
Road
Driving Maneuvers Surface Fz F~ Fy Mz
Standstill
Hydropuls Even <30
Steering Even
Starting Even <5 <5 Stat.
Straight line
Constant speed Even Stat. Stat.
Uneven < 30 < 30
Rough
Accelerating/braking Even <5 <5
Uneven < 30 < 30
Rough
Jolting Even <5
Shimmy Even < 30 < 30
Longitudinal ruts Even Star. Stat. Stat.
Cross-wind section Even Stat. <5 <5
/z-Step Even Stat. <5 <5 <5
Curve driving
Step steer input Even <5 <5 <5
Steady state circle Even Stat. Stat. Stat. Star.
Uneven <30 <30 <30
Rough
Power off/accel/braking Even <5 <5 <5 <5
Uneven <30 <30 <30 <30
Sinusoidal steer input Even <5 <5 <5 <5
/z-Step Even Stat. <5 <5 <5
a F~, vertical wheel load; F~, longitudinal force; by, lateral force; M~, self-aligning torque; Stat., static/stationary.
limit the number of input and output variables. The TYDEX Workshop tried to
find an optimum between practical use with a maximum of application possibilities.
Furthermore, the information about the road required by the tire model depends
on the tire model. A general purpose interface between fire model and vehicle
model implies also a definition of the interface between fire model and road model.
The proposed solution for a standard interface, developed by the TYDEX
Workshops, will be explained in more detail.
The tire parameters used in tire models should be derived from tire measure-
ment data. Tire test procedures depend largely on the type of application of the
tire model within the vehicle model. Differences in fire test results can be caused
not only by differences in tire test procedures but also by differences in tire test
devices. Even for a well-known fire property like cornering stiffness, rather large
inconsistencies can exist. In Fig. 1, results are shown of a survey by a group of
European vehicle manufacturers, tire suppliers, and research institutes on the com-
parability of test devices with respect to cornering stiffness [2].
van OOSTEN ET AL. ON MODELING, TESTING STANDARDS 191
1600
1500
1400
~ 1300
o) [~ Basis
~) 1200
"0. [---I--- 2,5 bar
1100
]--L~- 3 deg camber
1000
--O-- load 6000 N
9O0 I
I--X-- 140 km/h
800 I I [ I- I
2 3 4 5 6 7
Test bench number
FIG. 1 - - Comparison of cornering stiffness determined on seven different test devices. Basis." 60
km/h, 2.0 bar, no camber, 4 kN vertical load.
BMW AG Germany
Robert Bosch GmbH Germany
Centro Richerche FIAT Italy
Continental AG Germany
Daimler-Benz Germany
FHT Esslingen Germany
Ford Werke AG Germany
Goodyear S.A. Luxembourg
IPG GmbH Germany
Mercedes Benz AG Germany
Michelin France
NedCar The Netherlands
Porsche AG Germany
PSA France
Steyr-Daimler-Puch Austria
Volvo Car Corp. Sweden
Volvo Truck Corp. Sweden
TNO The Netherlands
Toyota (TCL) Japan
Univ. of Berlin Germany
Univ. of Delft The Netherlands
Univ. of Karlsruhe Germany
Univ. of Vienna Austria
Volkswagen AG Germany
files, abbreviations for the parameters, axis systems, and definitions are fixed
by the TYDEX-Format Manual.
The Structure of the Data Files. The TYDEX-Format was developed for
the filing of tire measurement data, but filing of tire model data is also possible.
An example of the file structure can be seen in Fig. 2. The data file is divided
in different sections having special keywords as a header.
The data file may consist of up to 12 keywords:
* *HEADER * *MODELPARAMETERS
* *COMMENTS * *MODELCOEFFICIENTS
* *CONSTANTS * *MODELCHANNELS
**MEASURCHANNELS **MODELOUTPUT
* *MEASURDATA * *MODELEND
**MODELDEFINITION **END
Column
41 51 61 71
**HEADER
RELEASE Release of T Y D E X - F o r m a t 1.3
MEASID M e a s u r e m e n t ID 05039ABC
SUPPLIER Data Supplier MICHELIN
DATE Date 20/02/97
CLCKTIME Clocktime 09:50
**COb~dENTS
This section can be u s e d to put in any comment. The format is free. B l a n k lines can be
used.
**CONSTANTS
N O M W I D T H Nominal Section W i d t h of Tyre mm 185
A S P R A T I O Nominal A s p e c t R a t i o % 70
TYSTHUCT Tyre S t r u c t u r e radial
RIMDIAME Nominal R i m D i a m e t e r inch 13
INFLPRES Inflation Pressure bar 2.5
INCLANGL Inclination A n g l e deg -3
A M B I T E M P Ambient T e m p e r a t u r e deg C 25
**MEASURCHANNELS
M E A S N U M B Measurement Point No.
RUNTIME Running T i m e s 0.01
FZH V e r t i c a l Force kN 0.001
SLIPANGL Slip A n g l e deg
LONGSLIP Longitudinal Slip % i00.
FYH Lateral Force N
FX Longitudinal Force N
MZH Aligning Moment Nm
**MEASURDATA
1 0. 4000 0.00 0.00 0. 0. 0.
2 i. 4000 0.02 -0.01 -200 -i00. 20.
3 2. 4100 I 0.04 -0,02 -400. 0. 40.
**MODELDEFINITION
MODELREF Cornering Stiffnesses DZ/GZ MICHELIN
**MODELCHANNELS
CSFYH C o r n e r i n g Stiff.Lat. Force N/deg out
CSMZH Cornering Stiff.Align,Moment Nm/deg out
FZH Vertical Force N in 500 10000
INFLPRES Inflation Pressure bar in 1.8 2.8
**MODELOUTFUT
221 5.869 500 1.8
1255 37.42 3000 1.8
3187 104.55 10000 1.8
**MODELEND
**END
3. If sections concerning the tire model are used, the keywords **MO-
DELDEFINITION and * *MODELEND must appear.
The meaning of the different sections beginning with the above-mentioned
keywords is explained in the following (see Fig. 2).
The first section is marked with the keyword **HEADER. This section
contains information to identify the measurement or model data. This is suit-
able for the file management. The section consists of only five parameters.
194 TIRE SCIENCE & TECHNOLOGY
I Ex I 2~
Abbrev. Parameter Unit Example CR
(8. char.) (max. 29 characters)
D. Ambient Conditions
TRCKTEMP temperature of track deg C 22
Fahrbab_ntemperatur
temperature de la chauss~e
HUMIDITY humidity % 60
Luftfeuchtiskeit
humidit~ de l'air
FIG. 4 - - The T Y D E X W-axis system; the f o r c e s a n d m o m e n t s act b y the tire onto the rim.
and institutions. Data transfer without problems is a requirement for this pro-
ject and can only be realized by a uniform data format such as the TYDEX-
Format.
The interface has been defined to be as universally valid as possible. Thus only
physically unambiguous parameters are exchanged, not those that can be defined
in different ways. Furthermore, it has been taken into account that the tire model
can obtain dynamic degrees of freedom that have to be integrated by the solver
together with the state variables of the vehicle model. In addition, no restrictions
have been made with respect to road properties, and it has been considered that
a vehicle can be equipped with an arbitrary number of different fires.
In this publication the interface shall not be described in detail. This is the
aim of the documentation that can be obtained at the contact addresses given
at the end of the article. The goal is rather to give a basic understanding of
the definitions.
Formally, the interface is the parameter list of a FORTRAN77 subroutine,
because FORTRAN is at the moment and will probably be during the next
years the standard programming language for external and user written sub-
routines in mulfibody simulation environments. It has been decided to have a
single precision version as well as a double precision version, which differ
with respect to their names and to the type of the parameters, but are, in
principle, identical.
Part of the input to the fire model are some control parameters, which state,
if the actual call to the model is during initialization, during normal simulation,
or if it is the final call to the model. Another parameter that is important for
dynamic tire models is the simulation time.
The next group of input parameters contains information about the kin-
ematic state of the rim. Values defining displacement, orientation, and ve-
locity of the rim are transferred in an unambiguous way. Using these pa-
rameters, the tire model can calculate internally such parameters as the
longitudinal slip, which can be defined in slightly different ways according
to the tire model.
The next group of input parameters are the state variables of the tire model
that have been integrated by the solver.
Furthermore, the tire parameters of the actually called tire are given to the
tire model. Here the interface is built in such a way that the parameters can
be transferred using the interface or can be read in from a file. Finally, a group
of parameters that has to be transferred to the road model follows.
The primary output of the tire model is the forces onto the rim at the center
of the wheel and the torques applied from the tire onto the rim. With this
definition all problems have been avoided that would occur if referencing, for
example, a "contact point" or a "contact area," which are very specific for
different tire models.
The next group of output parameters are the state variables of the tire model,
which have to be transferred to the solver.
In an info-array, any information about internal variables of the tire model
can be stored in the result file. The first part of this info-array is reserved for
van OOSTEN ET AL. ON MODELING, TESTING STANDARDS 199
arbitrary values like slip angle, longitudinal slip, or camber angle. The second
part of the array is up to users' needs.
Finally, the parameter list contains some work arrays for saving internal
values that are needed when the same tire is called the next time, and an error
parameter.
Besides the interface between the tire model and the vehicle model, a pro-
posal for an interface between the tire model and the road model has been
worked out. The basic assumption is that the road model or a special module
of the road model is called by the tire model. For a point (x, y), given by the
tire model, the road model has to deliver different information about road
properties, like the altitude z(x, y) and their derivatives at this point.
Both interfaces have been used in daily work by different members of the
TYDEX group for a long period, and all experiences have been incorporated
into the definitions. Thus the interface between the tire model and the vehicle
model has reached a very mature and stable state. The interface between the
tire model and the road model should be seen as a well thought-out proposal,
which, at the moment, might not cover all possible needs, but should be used
to gain further experience.
The present publication shall encourage colleagues all over the world to use
these interfaces in their own simulation environments. Detailed specifications
of the interfaces can be obtained at the contact points given in the Appendix.
Background
To develop vehicles that have maximum active safety, car manufacturers
have an increasing need for information about the force and moment properties
of tires. Vehicle manufacturers, tire suppliers, and automotive research organ-
izations have advanced test equipment to measure the forces between a tire
and a road surface under a variety of loading conditions. However, because
of the large differences in test equipment and the measurement procedures
used, the consistency of the tire force and moment properties determined with
the different test devices is a major problem. As mentioned in the section titled
Objectives and Approach, for an arbitrary tire property like cornering stiffness
one can find large differences when comparing this parameter obtained from
different tire test devices. Differences can be caused by differences between
the tire test devices, like surface friction, surface curvature, ambient condi-
tions, etc; inconsistencies between the measurement signal acquisition and data
postprocessing; or the measurement procedure itself, including warm-up.
The differences between the tire test devices must be seen as boundary
conditions, which are difficult to change. One could look for a correlation
between the devices by investigating the effect of the differences, but knowl-
edge about these effects and standardization on signal processing and the mea-
200 TIRE SCIENCE & TECHNOLOGY
surement procedures would improve the comparability of tire test data from
various test devices considerably, not forgetting that the starting point for tire
testing guidelines should always be the real life conditions of a tire during
usage on a vehicle.
In the TYDEX Workshop a discussion started on the development of guide-
lines for tire testing, resulting in the initiative to start a common research
project called TIME (Tire MEasurements, forces and moments) within the
scope of the Standards, Measurements, and Testing program of the EC (DG
XII) [ 5 ]. The consortium consists of the six major European tire manufactures,
three vehicle manufactures, four research organizations, and one commercial
automotive company.
Objectives
The TIME project aims at the development of a common tire measurement
procedure that will be reliable and consistent with realistic driving conditions.
The research will concern passenger car tires under steady state cornering
conditions (from low to high lateral acceleration levels up to the safety limits
of the vehicle).
The general objective will include the following items:
1. To set up a correlation between each of the main different tire test devices
and the reference test conditions.
2. To explain the differences in measurement results from different tire
testing devices.
3. To identify the conditions a tire experiences in real life usage.
4. To develop run-in and warm-up procedures for each test device before
testing a fire.
5. To establish the requirements for a tire testing laboratory (as a data
supplier) in terms of measurements quality assurance, testing procedure,
data acquisition, and processing.
6. To specify the validity range of tire data coming from a given procedure
with respect to actual driving conditions on a vehicle.
Approach
The activities of the project have been defined in five major workpackages.
In workpackage I, the differences between the results obtained with the main
European tire test devices will be investigated. Six different tire types will be
tested at each test device using exactly (if possible) the same measurement
procedure. Although the test program should be limited in order to compare
the most important tire properties, care must be paid to:
1. The run-in and warm-up procedures
2. The sequence of the tests
3. The selection of tire types and sizes
4. The number of tires required to execute the test program
van OOSTEN ET AL. ON MODELING, TESTING STANDARDS 201
Present Status
The TIME project started in February 1996 and will conclude with a stan-
dard tire measurement procedure in January 1999. At present the measure-
ments of Workpackage I have been performed and are evaluated resulting in
a program for Workpackage II.
faces: (1) an interface between vehicle models and tire models: the standard
tire interface; and (2) an interface between tire models and tire measurements:
the TYDEX-Format.
The STI is already available in the major commercial multibody simulation
software packages; the TYDEX-Format has proven its usefulness in the TIME
project and tire model related software.
Another result of the standardization efforts in the TYDEX Workshop is
the development of a standard tire measurement procedure within the Euro-
pean TIME project.
The participants of the TYDEX Workshop aim at a world-wide distribution
and use of their standards. A first review and evaluation of the TYDEX stan-
dards and the TIME standard tire measurement procedure is planned in 1999.
Until then, as much experience as possible will be obtained. This experience
shall lead to the next releases of the interfaces. All users of the interfaces are
cordially invited to influence the further development by sending their remarks
to the contact addresses and by taking part in the work of the TYDEX group.
They will be informed about the further process and will be invited to the next
meeting in the first half of 1999.
Appendix
Contact Addresses for TYDEX
General: TNO Road-Vehicles Research Institute, J. J. M. van Oosten, P. O.
Box 6033, 2600 JA Delft, The Netherlands. Phone: 31 15 2696420; Fax: 31
15 2697314.
TYDEX-Format: Universit~it Karlsruhe, Institut ftir Mkl and Kfz. Ban,
H.-J. Unrau, Kaiserstrasse 12, D-76128 Karlsruhe, Germany. Phone: 49 721
6083795; Fax: 49 721 6086051.
Standard Tire Interface: IPG, A. Riedel, P.O. Box 210522, D-76155 Karls-
ruhe, Germany. Phone: 49 721 9852013; Fax: 49 721 9852099.
References
[11 Zamow, J., "Overview of Typical Driving Manoeuvres and the Requirements on the Tire
Model," 2nd TYDEX Workshop, September 1992.
[2] Zamow, J., "Messung des Reifenverhaltens auf unterschiedlichen Prtifst~inden," VDI
Befichte hr. 1224, 1995.
[3] Unrau, H.-J., and Zamow, J., "TYDEX-Format, Description and Reference Manual," Release
1.3, Initiated by the TYDEX Workshop, 1996.
[4] Riedel, A., "Standardised Interface Tyre Model - - Vehicle Model," Release 1.3, TYDEX
Workshop, 1996.
[5] J. J. M. van Oosten, e.a., "TIME, Tire Measurements, Forces and Moments, proposal, Sci-
entific and Technical Content," Part A, TNO, Delft, 1995.