See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/342171454

Steering mechanism design for lightweight vehicle

Conference Paper · June 2020

CITATION READS

1 1,943

2 authors:

Zoltan Pusztai Péter Kőrös

Széchenyi István University, Gyor Széchenyi István University, Gyor
7 PUBLICATIONS   11 CITATIONS    33 PUBLICATIONS   49 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Electric Powertrain Development View project

Energy efficient BEV design and measurement View project

All content following this page was uploaded by Péter Kőrös on 15 June 2020.

The user has requested enhancement of the downloaded file.

 KORMÁNY MECHANIZMUS TERVEZÉSE KÖNNYŰ
SZERKEZETES JÁRMŰHÖZ
STEERING MECHANISM DESIGN FOR LIGHTWEIGHT VEHICLE
Zoltán PUSZTAI*, Péter KŐRÖS**
*Széchenyi István University Research Center of Vehicle Industry, Győr, Hungary,
pusztai.zoltan@ga.sze.hu
**Széchenyi István University Research Center of Vehicle Industry, Győr, Hungary, korosp@ga.sze.hu

ABSTRACT - In this paper the main design

concepts and calculations of an alternative
steering mechanism is presented, which is
dedicated for purpose-made lightweight vehicles.
Losses are generated during steering as the tires
are not rolling perfectly in curves, it’s caused by
lateral sliding. The widely used Ackerman
steering geometry has its specific error. In order
to avoid Ackermann error, the design of an
alternative steering mechanism was carried out.
This mechanism aims to keep both steered wheels
at ideal turning angles. This cannot be achieved
using the common rack and pinion steering, Figure 1. – SZEmission, the most recent SEM race car of
therefore a mechanism was designed based on our SZEnergy Team 2019
optimal spiral path method. The advantages of this
steering beside eliminating Ackerman error, is 2. GEOMETRY FUNDAMENTALS
that the mechanism is able to self-locking. This One of the most crucial sub-component of these
mechanism is ideal base for autonomous special vehicles is the steering system. The
application as the system doesn’t require energy steering geometry design largely influences the
to hold the turning angle, it just consumes energy vehicle dynamics. Ideal Ackerman geometry
when the angle is changed. This article covers the could be defined as a state, when the tire’s
main theoretical concept, design fundamentals, rotational axes are perfectly intersecting the center
FEM simulations and calculations. of the turn. [1]
This ideal geometry can be defined based on the
1. INTRODUCTION given vehicle data. The SZEmission parameters
The alternative vehicle development has great are shown in the table 1.
tradition in the Széchenyi István University in
Győr. The electric vehicle development team – Table 1. – SZEmission vehicle data
SZEnergy Team – has been participating in Shell Wheelbase (L) 1300 mm
Eco-marathon (SEM) since 2008, which is the Front track 1000 mm
world largest energy efficiency competition. Rear track 800 mm
Drag link radius (Ls) 19,5 mm
The participating vehicles have different Kingpin distance (b) 961 mm
propulsion system compared to pneumatic
Minimal turning radius (rho) 5676 mm
vehicles, but still has a lot in common. Our
development goals are to reduce the vehicle
mechanical, aerodynamical and electrical losses. These vehicle data were directly acquired from
In this article we’re presenting special steering the 3D CAD models of the car. The abbreviations
mechanism design, which is also applicable in can be found in the figure 2., which clearly shows
autonomous driving challanges as well. The most the main principle of Ackerman steering
recent battery electric vehicle of the SZEnergy geometry. Based on the presented kinematics
Team can be seen in the figure 1. scheme the following equations can be
established. The outer wheel angle can be
calculated from the vehicle data as follows. [2]
 Figure 3. – Characteristics of SZEmission’s wheel angle
There are methods when the wheels are turned by
Figure 2. – Ackerman steering geometry [2] computer controlled actuators to theirs ideal
𝐿 position. These steer-by-wire methods have no
𝑡𝑔𝛼 = direct physical connection between the steering
𝑏
R+ wheel and the rotating wheel, that’s why safety
2
(1) issues arise related to those systems.

Based on SEM regulations, we have to design a

The inner wheel angle can be determined physically connected steering system, which has
similarly as the outer. the lowest possible backlash and closest to ideal
𝐿 Ackerman geometry in our operating range. The
𝑡𝑔𝛽 =
𝑏 situation become more complex if we examine the
R−
2
(2) parts deformation under operating loads and
variable suspension geometry design with
different loadings. In this article we’re designing
The radius of curve depends on the wheelbase and unique steering mechanism, which is using
steering angle. optimal spiral path calculated from vehicle data
𝐿 with our method to guide the wheels according to
R =
𝑡𝑔𝛼 ideal Ackerman geometry. The calculations were
(3) carried out using simplified rigid suspension

The vehicle’s turning ability relates to the smallest

turning radius, which can be defined as the outer
front wheel’s center distance from the curves
center. The following equation describe the
minimal turning radius, marked as rho.
𝑏
𝑟ℎ𝑜 = √(𝑅 + )2 + 𝐿2 + 𝑙𝑠
2

(4) kinematic model. [2]

In case of Ackerman steering geometry, the inner 3. OPTIMAL SPIRAL PATH METHOD
and outer wheels are in different angle position. In order to apply the method, first we have to
The inner wheel angle (β) is always greater than create the movement function of steering shaft
the outer wheel angle (α), when we’re observing position compared to the rotation of the knuckle.
normal Ackerman geometry. The SZEmission’s The function’s directional correlation is shown in
optimal wheel angles depending on the actual the figure 4.
Figure 4. – Directional correlation in our movement
turning radius can be seen in the figure 3. function

Further on, we also used the previously presented

vehicle data table with adding other geometric
data such as base radius of our spiral geometry.
From CAD models we have to acquire the
appropriate data pairs of the ideal Ackerman angle
and the translational position change of our
steering shaft. Based on the calculations we graphically display
two rectangles in the figure 6. These rectangles
The value of base radius depends on the steering shows the range of 20 m turning radius and 30 m
torque, drag link radius and the expected turning turning radius from inside to outside respectively.
radius. Due to the modification of the base radius These are the most commonly measured turning
value we can achieve different force ratios in the radius values in a SEM race track.
steering mechanism. The listed parameters and
their abbreviations can be seen in the table 2.

Table 2. – Abbreviation for optimal spiral path calculation

Base radius for spiral path rbase
Steering shaft position – x direction xshaft
Optimal wheel angle λ

The optimal spiral path x and y coordinates can be

determined by applying the following geometry
transformation.
Figure 6. – Relation between steering shaft displacement
𝑥 = (𝑟𝑏𝑎𝑠𝑒 + 𝑥𝑠ℎ𝑎𝑓𝑡 ) ∗ cos(𝜆) (5) and wheel angle

𝑦 = (𝑟𝑏𝑎𝑠𝑒 + 𝑥𝑠ℎ𝑎𝑓𝑡 ) ∗ sin(𝜆) (6) It’s noticeable that the steering system mostly
needs to cover only 4 mm shaft displacement in
This x and y coordinates exactly describe the its operating conditions.
optimal spiral path. The rotating spiral path guides
the steering shaft translational movement which 3. STEERING MECHANISM DESIGN IN CAD
keeps the angles of the steered wheels in the ENVIRONMENT
position when their rotational axis perfectly CAD design implementation immediately started
intersect the center of the turning radius, after the calculations of the optimal spiral path
achieving the ideal Ackerman geometry. Due to were carried out. The most important part is the
this mechanism we can notice remarkable spiral disk. This part was design in two different
decrease in the suspension losses, particularly in concepts.
curves. The spiral path x and y coordinates are
illustrated in the figure 5. The points are In the one hand, the spiral path equation is drawn
connected with spline interpolation and the base in CAD with using spline tools and this curve is
radius is also shown in the figure. offset with given width. In the two sides of
resulted spline there are urethane molded
bearings, which are connected to the steering rods.
In this case the spiral guides the rods through
rolling bearings.

On the other hand, the disk can be designed with

doubling the extruded curve to create a hollow
between the curves. The inside of the hollow
could guide a properly fitted shaft. In both case
the rotation of the spiral disk is transformed into
the steering rods translation movement. The main
difference between these options is the way of
friction and the clearance between the parts,
which could easily cause undesired backlash in
the whole system.

The spiral disk and its shaft has to be supported by

Figure 5. – Spiral path x-y coordinates bearings from both side. We designed a framed
construction, which connects the bearing
The relation between the shaft displacement and housings. The frame and the housings are fitted
the wheel angle is almost perfectly linear as it is with positioning pins to keep the shaft in its
shown in the figure 6. designed place.
 as we had no exact data about the occurring loads.
The steering rods are connected to the spiral disk The load cases with different CAD data were
through adjustable arms. The rods are made from matched. The results of these cases were
hard anodized square sectioned aluminum bar. compared to each other to determine the trends.
The steering rods are supported and mounted by
plastics crossbars. The whole construction is We used the available basic ANSYS material
placed in U-profile aluminum frame, which unites models such as aluminum and structural steel.
parts and increases the system rigidity. The figure Tetrahedron meshing with local mesh tools in the
7. shows the CAD assembly of the designed spiral main contact area was chosen. We tried to make
path steering. mildly detailed mesh to check the models with
reasonable accuracy.

Some simulation input were received from CAD

design such as the geometry data of the bearings.
These commercial products have exact size
limitation, so we have to adjust our parts
dimension to them. We targeted to find
dimensions, which give the part near the same
strength as the bearings. It means that we didn’t
examined exact numeric results just compared the
Figure 7. – CAD assembly of the steering system different simulation cases. One of the static
structural simulations isoline results can be seen
SZEnergy Team is making great effort in in the figure 9.
autonomous driving development, that’s why an
autonomous steering extension was designed to
the presented steering system, illustrated in the
figure 8. The steering is driven by DC motor
through timing belt connection. The motor
controller gets position information from linear
sensor placed on the steering rod. The
autonomous extension is separate unit, so it could
easily attached to the steering system.

Figure 9. – Static structure isoline result

Evaluating the simulation results we agreed on 5

mm spiral curve width, which satisfy both
mechanical and manufacturing demands. The
manufactured and pre-assembled spiral disk can
be seen in the figure 10.
Figure 8. – Autonomous steering system

4. STATIC STRUCTURAL AND RIGID BODY

DYNAMICS SIMULATION IN ANSYS
The CAD design process was supported by plenty
of FEM simulation. Traditional static mechanical
simulations were made to determine the different
dimensions of the presented parts. Main goal was
to specify the width of the contacted spiral disk.
In case of static simulation we have to regulate the
system’s degree of freedom (DOF) with
constrains and apply loads to the system. We
made simulation cases with different sample loads
 Figure 10. – Pre-assembled spiral disk spiral disk is reached, reversed rotation starts to
Rigid body dynamics module of ANSYS was also take the simulation back to the initial condition.
used to receive approximate data of the system The simulation runs periodically. Joint and
movement characteristics. Firstly we had to position probes were queried to get the movement,
modify the simulation constrains and adjust the velocity (angular and translational) and force
system’s DOF to create a periodically repeated results. The comparison of the translational and
movement. The main simulation constrains and angular velocity is illustrated in the figure 13.
joints are illustrated with the simplified model in
the figure 11.

Figure 11. – Constrains and joints in ANSYS Rigid Figure 13. –Translational and angular velocity results from
Dynamics module ANSYS Rigid dynamics

The rotational and translational performance can

The sample spring load is similarly determined as be also calculated from the queried simulation
in the static structural module as the exact loads results as it is shown in the figure 14.
are unknown yet. The spring load connects the
fixed knuckle (simplified as a rectangular body)
to the steering rod. The steering rod can move
translationally in the supports. The polymer
crossbars are also fixed to the ground. No
separation contact is applied between the roller
bearing’s cylindrical surface and the designed
spiral path surface, the figure 12 shows that in
details.

Figure 14. –Performance results from ANSYS Rigid

dynamics

5. SUMMARY
Special steering mechanism concept based on the
calculation of the optimal spiral path was
presented in this paper. This spiral path creates a
unique curves depend on different suspension
characteristics. The curves follows the ideal
Ackerman angle in every position, which
extensively reduce the suspension losses. The
Figure 12. – No separation contact on the spiral path main design and simulation principles of the
steering systems were also presented.
The spiral disk is able to revolute around its local
Z axis. This revolute joint is driven with constant 13. BIBLIOGRAPHY
torque. The rotation is limited from 0° to 315°, it’s [1] N.W. Troupe.: Optimization of Steering Geometry For
Ultra‐High‐Mileage Vehicles, The Pennsylvania State
the physical limit of the spiral path. The University, 2011.
movement starts from the maximum distance [2] K. Hartani, Y. Miloud, A. Miloudi.: Electric Vehicle
from the knuckle. The spring is unloaded at the Stability with Rear Electronic Differential Traction,
beginning. The rotating disk moves the steering EFEEA’10 International Symposium on Environment
rod only in the global X direction, the rod Friendly Energies in Electrical Applications, Ghardaïa,
Algeria, 2-4 November 2010.
compress the spring which generates reaction
force in the knuckle. When the angle limit of the

View publication stats