Professional Documents
Culture Documents
Guidelines To Best Practices For Heavy Haul Railway Operations
Guidelines To Best Practices For Heavy Haul Railway Operations
Best Practices
For Heavy Haul
Railway Operations:
Management of the Wheel
and
Rail Interface
i
This book is the third in a series: Guidelines to Best Practices for Heavy Haul
Railway Operations
ISBN 978-0-911382-63-1
ii
The following chapter is excerpted from:
2.1 Introduction
The transverse shape and longitudinal running profile of wheel and rail are
essential to the effective guidance of a rail vehicle on the track. Wheel/rail
contact will change these shapes and profiles over time through wear,
deformation, and (in the limit) failure of the material on the running surfaces.
A change in shape can reduce the effectiveness of vehicle guidance. Material
failure of the running surface compromises the structural integrity of the wheel
and/or rail. Compromised vehicle guidance and failure of the running surface
will accelerate vehicle and track degradation.
The task of vehicle and track designers is to provide cost-effective products
that reduce wheel and rail forces, stresses, deformations, and wear, retaining
desirable wheel and rail shapes as long as economically possible.
The task of railroad operators is to choose appropriate designs for their
operating circumstances and maintain vehicles and track, and in particular the
contact interface, within limits to provide for safe and cost-effective operations.
Operations can differ significantly from one railroad to another as they can
have very different parameters such as track topology (curvature and grade),
axle and train loads, gage, climate, geology, operations, and past operational his-
tory. Any one of these parameters can significantly influence the many choices
of wheel/rail interface design or maintenance practices used. Consequently, no
single “recipe” for success exists as there is a large list of possible “ingredients.”
An informed understanding of the basic mechanisms of wheel/rail and vehicle/
track interaction can assist in choosing products and strategies and avoiding
potential conflicts. The object of this chapter is to describe these basic mecha-
nisms and their impact on wheel/rail interaction. This description will be qualita-
tive and should help the reader to make informed decisions on product choice
or maintenance intervention. Quantitative answers are best left to experts in the
many different fields involved.
Note that the IHHA published a handbook on wheel/rail interaction in 2001.2.1
In the ensuing 14 years, the industry knowledge base has broadened and more
tonne-kilometers (ton-miles) are being operated at elevated loads. Wagon and
bogie design and track maintenance practices are evolving to suit.
The science of contact mechanics under heavy haul operations is being
increasingly better understood, particularly the role of steering tractions in
the development of rolling contact fatigue. This has led to the development
and implementation of improved steering vehicles and bogies and the increas-
ing application of contact-conditioning technologies such as top-of-rail friction
control. The track has been strengthened and wheel profiles and grinding
practices implemented to counter rail rollover. Wheel steels are being im-
proved and wheel machining practices updated to counter the incidence of tread
shelling and wheel failures such as vertical split rims. Rail materials and
cleanliness are continually being upgraded.
2-1
This chapter has been revised to reflect these changes. Heavy haul
development continues and the reader should expect further developments in
the coming years, particularly in the science of contact mechanics (reference
chapter 3).
Some of the typical questions that may be answered in this chapter are:
• How does a wheelset/rail vehicle steer on the track?
n What is the importance of wheel and rail profiles?
n What is conicity?
• What is vehicle hunting?
• Is there a compromise that must be made between vehicle curving and
hunting?
• What significance do pitch, bounce, and rock and roll modes have in
vehicle/track interaction?
• What generic forces are generated between the wheel and the rail and
what damage might these forces do to the track? (reference also
chapter 3)
• Considering the basic mechanisms of wheel/rail and vehicle track
interaction:
n What are the main system parameters I must consider when designing
a heavy haul operation?
n What are the important maintenance parameters I can control?
2-2
The track comprises two rails laid on sleepers at a particular gage, as
figure 2-2 shows. The rails are laid at an angle, β, to the sleeper to generally
match the angle, γ, of the wheelset profile. This centralizes contact on the center
of the head of the rail on tangent track and assists in stabilizing the rail against
rollover as the normal reaction to the contact as the wheel generally passes
through the base of the rail.
2-3
some wheels and the rail if excessively stiff vertical suspension designs are used.
Furthermore, the running surface of the rail is discontinuous at nonwelded rail
joints and certain types of crossings at turnouts. This can cause impact loads on
the wheel and the rail and, momentarily, result in a shift in the wheel/rail contact
position on the wheel profile. This may affect wheelset guidance.
Steel-on-steel contact produces a uniquely low rolling resistance for railway
vehicles. The geometry of the wheelset, described mathematically as a di-cone
(two cones placed back to back having a cone angle, 2γ), imparts on the wheelset
unique properties of self-guidance (i.e., self-centering on tangent track as figure
2-3 shows and the ability to negotiate curves as shown in figure 2-4).
The di-cone design also provides the railway wheelset with the ability to
accommodate diameter inaccuracies between the two wheels by displacing lat-
erally on tangent track to compensate for the diameter difference. The flange
on the wheel tread is used where track discontinuities or track geometry
irregularities are so severe or the vehicle suspension is so inadequate that the
properties of self-guidance of the wheelset are insufficient for guidance without
flange contact.
2-4
2.3 Kinematics of a Railway Wheelset
2.3.1 Introduction
The dynamic interaction between the vehicle and the track, and thus between
the wheel and the rail, has a significant influence on the running safety and ride
quality of railway vehicles as well as on the fatigue and wear of the wheel and the
rail surface. It is thus essential to have a good understanding of the impact of,
for example, heavier axle loads on critical system performance issues and their
impact on safety and system degradation.
The wheel tread is generally conical or has a particular curvature in the
transverse direction (described in section 2.2). Figure 2-5 shows the wheel tread
and other typical features of a wheel profile.
2-5
2.3.2 Rolling Radius Differential
The difference between the rolling radii on a wheelset is known as the “rolling
radius difference” and is an important parameter in analyzing the behavior of the
railway wheelset.
Figure 2-7 shows the rolling radius difference, Δr, as defined by:
Δr = γ y (2.1)
where y is the lateral displacement of the wheelset from the center position of
the track and γ is the conicity of the wheel tread. For conical wheels, the conicity
is equal to the contact angle of the wheel profile.
3 N P₀,max =
(Note that the figure includes Redtenbacher’s formula2.3P₀(ξ,η)
) = 2 Ae √ 1 - ( aξ )² - ( ηb )² ,
Figure 2-7: Rolling of a Coned Wheelset on a Curve
(32) =
τ₁,max = 0.3 ∙ p₀ at z = 0.78a.
2.3.3 Conicity
For wheelsets with conical wheels, the conicity is derived from the mean slope
of the rolling radius versus lateral wheelset displacement and
Rw δis defined as: r0 l
Δr γe = y =
(2.2) γ = y R w - RR Rc γ
For circular wheel/rail contact geometry (figure 2-8 is an example) a so-called
“equivalent conicity” is derived from equation (2.2) in terms of the profile radii:
f E
kb² = 11 kp l²
f22
Δr = r0 R + l - 1
R-l [ ]
Δ r γ y R R δ2l λ π w e
lr
(2.3) Rw δ r l λ = 2π 0
γe = y = 0 γ
γ = Δr R - R y w R Rc γ
Where Rw is the wheel profile radius, RR is the rail profile radius, and δ is the
angle between the plane f11of contact and track level.
2.1, 2.2
Effective Conicity (λ) = R
reduces to:
kb² =
When Rw is set to infinity
f22
k l²
and
p Rw
is smallΔr =
(ar R+l -1
[ ]
0typical rail profile), equation (2.3)
R-l
2
γe = δ (2.4)
or the angle of the wheel profile. γ = rolling radius
2y
unconstrained massless wheelset not subject to any external forces. The “pure
rolling” position thus describes a “zero force” state of the wheelset. Deviations
rolling radius differential
from this position result in the generation of forces between γ =wheel and2yrail.
On curved track, the outer wheel of a wheelset has a greater distance to travel
than the inner wheel. To compensate for this, the wheelset thus moves laterally
relative to the track so that the larger wheel radius on the inner side of the wheel
runs on the outer rail of the curve. The inner wheel uses the outer side of its
tread to reduce the traveled distance during the passage through the curve. This
causes each wheel to run a different distance per revolution of the axle. Hence,
the wheelset alignsΔ r γitself
y R Rradially
δ2l λ π in curves. e
w R
2-7
2.3.5 Self-Centering Action
The geometry of the wheelset allows it to negotiate curves due to the radius
differential generated between the wheels on the common axle, as well as the
unique property of self-guidance [i.e., self-centering on tangent track (figure
2-9)].
2-8
Hunting is an oscillation, usually unwanted, about an equilibrium position.
The expression came into use in the 19th century and describes how a system
“hunts” for equilibrium. A classical hunting oscillation is the yawing motion of a
railway vehicle caused by the coning action on which the directional stability of
a3 railway
N depends. P₀,max = 32 - NA
P₀(ξ,η) = 2 Ae √ 1 - ( aξ )² - ( ηb )² , e
In 1883, Klingel gave3 the firsttan(δ)
2.5
mathematical analysis of the kinematic
oscillation of an
τ₁,max = 0.3 ∙ p₀ at z = 0.78a.
unconstrained (
(2) = wheelset
1 + μ *
) -μ
and
tan(δ) derived the relationship between
the wavelength λ and the wheelset conicity γ, wheel radius r0, and the lateral
distance between 3contact N points between wheels and rails 2l as:
P₀,max = -
√ 1 - ( aξ )² - ( ηb )² , e 2 A
lr v
R
wδ (2.6)
l -μ λ = 2π 0 f = 0
()=( )
r
tan(δ)
γe = 3 y = 0 γ λ
R w - RR
₀ at z = 0.78a.
2 Rcμγ* tan(δ)
1+
According to Klingel’s formula, the frequency of the kinematic oscillation will
increase with increasing speed. The frequency of the kinematic oscillation is
defined as: Effective Conicity (λ) = RRD
p l²
RR
y = 0 [
Δrr =l r0 R + l - 1
R - l ]
(2.7) λ = 2π
lr0
γ
f =
v0
λ
2y
Rc γ
Lateral oscillations caused by coning were experienced from the early days
of the railways. One solution to the oscillation problem that has been proposed
γ = rolling radius differential
from time to time, even
Effective down
Conicity to RRD times, was
modern 2y to fit wheels with cylindrical
= r0
[ R+l -1
] (λ) =
2y
R - l treads. However, in this case, if the wheels are rigidly mounted on the axle, very
slight errors in parallelism would induce large lateral displacements that would
be limited by flange contact. Thus, a wheelset with cylindrical treads tends to
run in continuous flange rolling radius
γ = contact. It isdifferential
probably for this reason and the resulting
accelerated wear that cylindrical wheel 2y treads have never been used successfully
in heavy haul service.
Δ r γ y R R δ2l λ π w e
2.4 Quasi-Static Forces Acting on a Railway Wheelset R
√
2.4.1 Introduction
R δ2l λ π In this section,e
the forces acting in the wheel/rail contact are introduced.
w
These forces, to a great
R
extent, determine the running behavior of the vehicle.
√
They are, however, equally important to understand the damage mechanisms on
wheels and rails introduced later in chapter 3. For the sake of simplicity, only the
quasi-static situation is described where dynamic oscillations of the vehicles are
not taken into account.
2-9
Figure 2-10: Creep Force as Function of Creep
The following interesting observations can be made:
• When creep is zero, the creep force is zero. In other words, to develop a
creep force, creepage in the wheel/rail contact is needed. This means
for example, to develop a traction or braking force, longitudinal creep in
the wheel/rail contact is needed.
• For small creepages, the relationship is linear. The creep force, however,
saturates at the value μN (i.e., this is the highest value of creep force
possible to transfer in the contact).
• Given knowledge of the creep relationship (example figure 2-10), if the
position of the wheelset on the track is known, the creep may be
calculated together with the creep force.
The creepages and resulting forces acting across the contact patch can be di-
vided into three components:
• Longitudinal creep and resulting force
• Lateral creep and resulting force
• Rotational creep around the axis normal to the contact patch, which is
called spin or spin creep and results in a reacting moment and coupled
lateral force
The direction and magnitude of these creepages and creep forces are explained
in the following paragraphs.
2-10
into a position at right angles to the track center line by forces FJ applied by the
vehicle suspension to reach the final actual state shown in figure 2-11. This is
only possible if the wheel rolling on the larger radius slips backward and the
smaller wheel slips forward. This slip induces forces FS.
2-11
2.4.2.2 Lateral Creep
Lateral creepage can be explained in a similar way. For a wheelset placed at an
angle of attack α, the angle between the wheelset axle and a line perpendicular
to the track center line is shown in figure 2-13. A pure rolling wheelset would
roll off the track to the preferred final state indicated by the chain-dotted lines in
figure 2-13. As the wheelset is constrained by forces FJ acting at the journals (as
shown in figure 2-12) and by possible flange forces, it has to stay on the track.
This is only possible with a slip motion, directed to the outside of the curve, if
the wheelset moves forward with velocity V. The amount of lateral creepage is
proportional to the attack angle of the wheelset against the track.
2-12
2.4.2.5 Gravitational Forces
The wheel/rail contact also has to transfer the weight of the vehicle into
the track structure. The sum of the vertical forces Q is equal to the axle load
(figure 2-14). Because of the profiled wheels, the contact patches are inclined
at a general contact angle γ. This results in the generation of a lateral compo-
nent to the vertical load at each contact patch. When the wheelset is placed cen-
trally on the track, the lateral components at each contact patch are equal and
opposite. However, this situation changes when the wheelset is displaced from
the track center (figure 2-14). The angle of the contact patches on each wheel can
be different and result in a net force acting on the wheelset.
2-13
Figures 2-15 and 2-16 show typical forces occurring when flange contact
occurs.
Figure 2-17: Forces on a Wheelset in Flange Contact and Resultant Wheel and Rail
Damage
2-14
2.5 Introduction to Damage Criteria
Wheel/rail forces between the vehicle and track are transmitted across the
contact patches. They create stresses and creepages across these contact
patches and can generate one or a combination of the following at and/or
adjacent to the contact patch:
• Wear
• Material flow
• Surface and subsurface flow
The science of prediction of these parameters in the railway environment is
still in a developmental stage. The status of this science is provided in chapter 3.
Vertical loads and stresses are driven by axle load and vertical dynamics. These
are managed through load control, minimization of vertical dynamics, and wheel
and rail profile design and control to maximize the area of the contact patch and
minimize contact stresses. Wheel profile design is addressed in section 2.7.
Lateral loads are produced by the steering properties of the vehicle and are
a current strong driver of heavy haul bogie design, particularly on moderate to
severe curvatures and in severe, cold-climate conditions. Consequently, much
emphasis in this chapter is placed on bogie design and steering properties
(reference section 2.6).
2-15
2.6.1 Vertical Suspension Considerations
The purpose of the vertical suspension is threefold:
A. Attenuation of Vertical Vibrations
When moving forward on the track, the vehicle experiences vibrations of
varying frequencies. These vibrations excite the various modes of the vehicle
structure, body, and the payload. The dynamic body modes are: Bounce, roll,
pitch, nosing, and sway, as well as coupled modes of these basic motions. Some
of the mechanisms exciting the wagon body are:
• Long wavelength, track alignment irregularities in the vertical profile and
track twist. These irregularities typically result in vehicle input
frequencies between 0.5 and 30 Hz.
• Long wavelength, track stiffness variations are also present and activate
the vehicle in similar modes and frequencies as the alignment
irregularities.
• Short wavelength, consistent stiffness variations associated with local
sleeper support, result in vehicle input frequencies up to 40 Hz.
• High-frequency impacts at rail discontinuities (P1 forces) often excite the
vehicle body vertical modes to induce the so-called P2 lower frequency,
reaction forces.
B. Equalization of Wheel Loads by the Vehicle Suspension
As the vehicle is supported on a minimum of four contact patches (two wheel-
sets) on perturbed or twisted track, it is generically a statically indeterminate
structure similar to a four-legged table on an uneven floor. As section 2.2.1
states, sufficient vertical load is required across the contact interface for effective
guidance. The vertical suspension stiffness must thus prevent unacceptable
wheel unloading on twisted track.
C. Attenuation of Vertical Vibrations to the Track Structure
As a result of vertical, vehicle dynamics, dynamic loading is transmitted from
the vehicle through the wheel into the track substructure. Track elements, such
as the rail, the rail pads, the sleepers, as well as the ballast and the subballast
layer, are thus directly influenced by the dynamic performance of the railway
vehicle.
Typical exciting mechanisms are:
• Vehicle body dynamics in the frequency range between 1 and 30 Hz
• Out-of-round wheels (10 to 20 Hz)
• Wheel flats (10 to 20 Hz)
• Rail irregularities, such as rail joints and rail burns
Constraints that challenge the vehicle dynamist to design the optimum vehicle
suspension are typically:
• Limit on the minimum vertical, vehicle stiffness because of a limit on the
coupler height differential between adjacent vehicles in the tare and
loaded condition
2-16
• Volume occupied by, as well as the stresses within, the suspension
• Coupled dynamic modes, particularly those of:
n The coupled excitation of wheelset and the wagon body (one inducing
resonance of the other)
n The excitation of the wheelset and the wagon body by regular track
features, particularly rail joints and deformation of the track structure
åßinduced by the action of the vehicle at rail joints
• Initial cost of the suspension
• Maintenance cost of the suspension
Similarly, the track engineer is limited in what can be done to optimize the
track structure. Typical constraints are:
• The cost of track components (example: ties, rail pads, under-tie pads)
• Cost constraints on the amount of ballast and formation material and
other geotechnical materials that could be used (example: ballast mats,
micro-piles, rammed aggregate piers, etc.)
• Track construction and track maintenance costs
Track dynamics analysts often include the so-called Hertzian stiffness in
their analysis. This is the vertical stiffness attributed to the deformation of the
wheel and rail under load. It is a high-order stiffness and is associated with
high-frequency vibrations and impacts. These are mainly of concern to track
engineers, and hence this effect is included under the section on track support
structures.
In its simplest form, the suspension of a railway vehicle comprises four spring
assemblies vertically coupling the four journal bearings on two wheelsets to the
body (see figure 2-18). The four springs can be designed to fit within the space
for relatively light vehicle loads. However, as the vehicle becomes heavier and
larger, the following factors come into play:
• The ability to accommodate track twist by means of spring deflection
clashes with the demands on coupler height differential limits between a
loaded and an empty vehicle.
• The available space for springs and dampers in the region of the journal
bearing is limited.
• As the carrying capacity of the vehicle increases, the wheel base
increases. This leads to increased demands on vertical deflection to
accommodate track twist.
2-17
Figure 2-18: Simple Vehicle Suspension Arrangement
2-18
2.6.2.1 The Rigid Frame Bogie
The rigid frame bogie acts, vertically, very much like the model of the simple
railway vehicle described above. As indicated by the name, the single, rigid bogie
frame is typically in the form of a rigid “H” shape as figure 2-20 shows. The load
of the vehicle body is transferred from the center pivot through the “H-frame” to
the springs placed above the journal bearings. These springs form the vertical
suspension and cater to all the requirements for the suspension as listed above.
This type of bogie has not found favor in heavy haul operations for the follow-
ing reasons:
• Space constraints for springs with adequate carrying capacity and
deflection in the region of the journal
• Cost of providing four separate spring/damper systems on the bogie
• Cost of the H-frame from a manufacturing complexity and tolerance point
of view
The three-piece bogie has proven superior in all these respects.
2-19
• Suspension springs are in a region of the structure where more space is
available than at the journal boxes.
2-20
Consequently, three-piece bogies are often characterized by:
• Poor curving performance. As figure 2-22(a) shows, the bogie not only
constrains the wheelsets parallel to one another (as opposed to
permitting radial alignment of the axles) but also, through the warp
deflection of the side frames, sometimes allows the wheelsets to take up
exaggerated angles of attack. This induces high curving forces, flange
wear, and rolling contact fatigue on the wheel treads.
• Poor hunting stability. Figure 2-22(b) shows poor hunting stability as the
wheelsets may not be held “square” to one another in a sufficiently stiff
manner.
Despite these drawbacks, three-piece bogies continue to be the most com-
mon type of bogie design for heavy haul freight wagons. However, over the past
two or three decades, researchers and vehicle designers have developed a far
better understanding of vehicle dynamics. They now have the ability to develop
accurate computational models that has led to the development of a number of
advanced bogie designs.
2-21
Figure 2-23: Model of Vehicle/Track Coupled Dynamics for Freight Car
with Three-Piece Bogies (end view)
2-22
The axle box suspension is modeled typically as shown in figure 2-25.
2-23
Figure 2-27: Wheel/Rail Coupling Model
Utilizing the vehicle/track coupled model, the dynamic behavior of the en-
tire vehicle/track system can be analyzed, including vehicle hunting, curving
performances, and vehicle/track interactions.
Figure 2-28: Lateral Displacements of the Wheelset for Speeds Relative to the
Critical Speed
Figure 2-28 is, however, only valid for a completely linear system. In reality,
several nonlinearities are in the railway system, both regarding the wheel/rail
contact kinematics and mechanics and the vehicle suspension. The real hunt-
ing behavior commonly occurring can rather be described by a so-called limit
2-24
cycle diagram for nonlinear systems (figure 2-29). For stability studies of vehicle
systems in practical engineering, the nonlinear critical speed vB is generally used
as an important index to evaluate the vehicle performance.
2-25
2.6.3.3 Hunting Characteristics of Empty and Loaded Wagons
To further understand the effect of wagon load on hunting characteristics,
note that the conventional three-piece bogie utilizes the suspension wedges to
provide a stabilizing coupling between two coupled wheelsets. For three-piece
bogies using load-sensitive friction damping, the bolster suspension contains an
inner coil spring and an outer coil spring with different free heights, respectively
(figure 2-30).
For the empty wagon, only the outer coil spring is compressed with a small
stiffness and large static deflection. For the case of the loaded wagon, as the load
increases, the outer coil spring is being compressed gradually to the position
of the free height of the inner coil spring first (first stage), and then both the
inner and outer coil springs bear the load together (second stage). This results
in a bilinear force characteristic of the suspension as shown in figure 2-31. The
symbols f1 and f2 are the static deflections for the empty and the loaded wagons,
respectively. fT denotes the static deflection at the transition from the first stage
(stiffness K1) to the second stage (stiffness K2). In general, the stiffness K1 rep-
resents the load capacity for the tare (empty) condition, while the stiffness K2 is
for the loaded wagon condition. Meanwhile, the larger vertical stiffness of the
coil spring in the loaded condition causes an increase of the lateral and warp
stiffness, enhancing the elastic constraint capacity of the side frame to the bol-
ster, thus improving the running stability of the three-piece bogie.
2-26
Friction damping in the spring nest contributes the main damping effect
(figure 2-32). According to the kinematics of the components, two different
working states exist for the damping system (figure 2-33). From empty to loaded
conditions in both states, the normal forces (F1, F2) in friction pairs, as well as the
support force (P), increase, leading to larger lateral and vertical friction forces.
The contact friction further strengthens the damping effect, which could restrain
the hunting oscillation.
Figure 2-32: Relative Motion between Side Frame, Wedge, and Bolster
2-27
flange contact is shown in figure 2-34. The “clockwise” moments resulting from
longitudinal creep are in balance with the “anti-clockwise” moments resulting
from lateral creep. Hence, the bogie is kept in equilibrium.
2-29
Figure 2-37: The Frame-Braced Truck
2-30
Figure 2-39: Adapter Pad Fitted to a Premium Bogie
This feature provides some freedom for the wheelsets to align:
• Into a parallel position relative to one another to compensate for
manufacturing tolerances in the bogie
• Into a partially radial position in shallow curves to improve curving
2-31
2.6.4.2.1 Interaxle Shear Stiffness
Research has shown that a certain amount of interaxle shear and bending
stiffness is required for adequate dynamic vehicle stability.2.9, 2.10, 2.11, 2.12
Conventional bogies with a rigid bogie frame use the shear stiffness of the bo-
gie frame to obtain interaxle shear stiffness. However, in the three-piece bogie
arrangement, the shear stiffness of the three-piece frame is inadequate for
optimal stability. Furthermore, if the frame is used for the transmission of shear
forces between the wheelsets, a high wheelset yaw constraint is required. This
is contrary to the requirement of steering bogies.
Consequently, to optimize the curving and stability performance of three-piece
bogies, interaxle shear stiffness, independent of the wheelset yaw constraint,
is required. In this case, the longitudinal and lateral stiffness of the primary
suspension can be selected to best suit curving performance, stability, and ride 3
quality. 3 N P₀,max =
P₀(ξ,η) = 2 A √ 1 - ( aξ )² - ( ηb )² ,
e
2
where
k = longitudinal axle box stiffness
2b = distance between axle boxes
f11 = longitudinal creep coefficient
f22 = lateral creep coefficient
kp = profile stiffness
2l = distance between wheel/rail contact points
If this relationship is adhered to, high, interaxle, shear stiffness can be
provided for optimal hunting stability without adversely R R δ 2 l λthe
Δ r γ yaffecting π off-flangew e
curving performance of the bogie.2.11
2-32
2.6.4.3 Developments in Self-Steering, Three-Piece Bogies
The abovementioned concepts have led to the development of self-steering,
three-piece bogies such as the cross-anchor bogie, the bissel (or wishbone)
frame bogie, and the radial-arm bogie. These bogies are briefly described below.
2-33
2.6.4.3.3 Radial-Arm Bogie
The radial-arm bogie (figure 2-43) is a further development of the cross-anch
or bogie. Like the cross-anchor bogie, it connects the two wheelsets of the bo-
gie in shear. This is not achieved, however, by cross anchors, which have to be
fitted diagonally through the bolster and be connected to subframes, but by ra-
dial arms that are positioned at the outside of the side frames. For three-piece
bogies, the radial steering arms form an integral part with the bearing adapters
that rest on the package-type journal bearings.
2-34
• Similarly, there are mechanisms that create the desired wheelset
alignment by reacting to the lateral displacement of the vehicle body
relative to the truck frame when traversing a curve. Again, this style of
mechanism depends upon the unbalance in a curve to create the steered
alignment, and at any other than balance speed in a curve, the steering
will be incorrect.
Both mechanisms mentioned above also suffer from the fact that any roll
or lateral disturbance experienced in tangent track will create a steering
response, which could be detrimental to the vehicle’s behavior.
• Another class of mechanism that has been proposed utilizes sensors and
powered actuators to achieve the desired steering effect. The curvature
of the track is detected by a sensor on the truck, this sends a signal to an
actuator, and the actuator moves the wheelsets into the desired position.
• A further type of body-steered truck has been developed in which the
degree of curvature of the track is determined by the yaw angle between
the bogie frame and the vehicle body. The yaw angle of a bogie in a wagon
is directly proportional to the degree of curvature of the track. By
connecting a mechanism between the truck frame and the vehicle body,
which moves in response to this swivel angle and acts upon the wheel
sets, it is possible to effect a displacement of the wheelsets relative to the
bogie frame such that the wheelsets must always be in perfect radial
alignment with the curvature of the track. Such a mechanism is not
dependent upon the speed of the vehicle – the yaw angle is independent of
that factor – so the steering will be the same for all conditions of over-
balance and under-balance speed in a curve. This type of mechanism also
requires no external power source or electronic device in order to be
effective. Even given this definition of steering, it is possible to have two
distinct types of bogies – ones in which the wheelsets are constrained
from moving relative to the bogie frame, and thus the mechanism must
overcome this constraint – and ones in which there is no constraint against
the angular motion between the wheelsets and the truck frame, so the
mechanism simply guides the wheelsets into the correct position and
constrains them to stay there. A schematic diagram of a body-steered
bogie is shown in figure 2-44.
2-35
Figure 2-44: A Steered-Axle Railway Truck
This mechanism allows both axles to be steered to equal and opposite angles
relative to the transverse center line between them and that there must be little
or no lateral (shear) motion between them. With this condition met, the wheels
will follow one another on the same path through a curve.
There are a number of design considerations for a body-steered bogie:
• The axles must be able to move freely relative to the bogie frame for
radial alignment under the vertical load at the bearings.
• The introduction of steering errors by the mechanism chosen must be
avoided. The yaw angle of the bogie under the wagon body is used as the
input to determine the amount of yaw of the axles. There are
nonlinearities in the system that need careful design.
• A coupling between wheelset yaw and the wagon body yaw could
introduce instability.
When negotiating a transition curve or spiral into a curve, the input to the
body-steered system will not reflect the true local radius of curvature at each
of the bogies, so there will be errors in the steering. The leading truck will be
under-steered and the trailing truck over-steered until the vehicle is fully onto
the simple curve.
These considerations can be overcome. Tests indicate steered-axle bogies can
reduce rolling resistance on tangent as well as curved track, improve hunting
stability, and improve derailment stability.
2-36
changes in shape, suggests that any change in shape introduces deleterious
conditions, and offers means to best minimize and manage this change in shape.
(a)
(b)
(c)
Figure 2-45: Typical Wheel/Rail Contacts under Normal Conditions
Variations in (a) wheel/rail profile designs, especially in grinding tem-
plates, (b) actual ground rail shapes, (c) wheel/rail wear, (d) track geometry
perturbations, and (e) vehicle performance can cause more (and more complex)
actual wheel/rail contact conditions in service.
2-37
Wheel/rail contact may also be described in the manner shown in figure 2-46
and related to three main regions on the wheel tread and rail head.2.22
Where R is the curve radius measured from track center line, l is half distance
of two rail contact points (see figure 2-7), and r0 is the nominal wheel radius.
γ = rolling radius differentia
Conditions for the onset of on-flange curving depend on the curving perfor-2y
mance of bogies, track gage, and wheel/rail clearance.
2-38
2.7.2.1 Wheel/Rail Contact Conditions in Tight Curves
(a) Strong, two-point contact (b) Strong, one-point contact (c) Conformal contact (d) Near-conformal contact
Figure 2-47: Contact Conditions on High Rail
Of these contact conditions, the strong, two-point contact and the strong
one-point contact with small contact area are not desired.
Strong, two-point contact has one contact point on the wheel tread and another
on the flange. A rolling radius difference is developed between these two points
that can reach 12 mm (0.47 in.) or more. This results in:
• Longitudinal creep forces acting in opposite directions, as figure 2-48a
shows. As a result, the wheelset steering moment is reduced, resulting in
higher angle-of-attack and lateral forces.2.23, 2.24
• Relative slip between these contact points and resulting high rate of wear
and material flow on the flange tip (figure 2-48b).
(a) Opposing Longitudinal Forces Cause Reduced Steering Moment (b) Plastic Flow at Flange Tip Contact
Figure 2-48: Detrimental Effect of Strong Two-Point Contact
2-39
The causes of strong, two-point contact are generally:
• Incompatible wheel and rail profile designs
• New wheels contacting the worn high rail
• Wheels with high flanges contacting excessively worn rail (figure 2-49)
• Overgrinding the rail gage corner (figure 2-50a)2.1
• Recanting rails that have rolled on their rail seats in service without sub
sequent rail grinding (figure 2-50b)2.25
High Rail
Figure 2-50: (a) Over-Cut Gage Corner (b) Recanting Rail without Rail Grinding
Single-point contact generally results in a small contact area at the high rail
gage corner and lower lateral forces because of a high rolling radius difference
between the wheels on the high and low rails and the resulting gravitational
force. However, high longitudinal creepages developed as a result of the high
radius differential Δr, in conjunction with high-contact stress areas, can cause
either, or both, a high rate of wear or surface cracks due to rolling contact
fatigue. Figure 2-51 shows an example of how the wear of a new high rail under
high-friction conditions (no gage face lubrication) resulted in the wear pattern
shown. The new rail profile was subjected to a strong one-point contact with
resulting high stresses and gage face cracking. High longitudinal tractions
caused a high rate of gage face wear until the wheel and the rail stabilized in
conformal contact conditions. Initial gage face lubrication, with an improved
conformal “match” of wheel and rail profiles with lower contact stresses allowing
the lubricating film to withstand the contact stresses, would have appreciably
reduced wheel and rail wear.
2-40
Figure 2-51: Wear Pattern Producing Strong,
One-Point Contact
Near-conformal and conformal contact on the high rail are more desirable
contact situations to produce lower lateral forces, rolling resistance, and contact
stress. They can be achieved by:
• Designing compatible wheel and rail profiles
• Regular rail grinding and wheel/rail natural wear
Under heavy haul operation conditions, wheel and rail profiles that are
designed for near-conformal contact soon wear into full-conformal contact,
resulting in extended wheel and rail reprofiling intervals.
2-41
Contact at low rail gage corner (figure 2-53) can reduce the rolling radius
difference Δr on the wheelset, resulting in flange contact and higher lateral
forces. It can cause material flow on the low rail gage corner, reducing the gage
clearance and moving the contact position further to the low rail gage corner.
This contact condition can occur:
• When excessive material is removed from the field side of the low rail
• Under conditions of low rail roll
Figure 2-53: Contact toward the Gage Corner on the Low Rail
Contact on the field side of the low rail (figure 2-54) can:
• Reduce the rolling radius differential Δr on the wheelset, depending on
the (worn) shapes of the wheels
• Result in high-contact stress due to small contact area and cause metal
flow on the field side of both wheel and rail
• Cause the rail to roll if not adequately constrained (example: if secured
with cut spikes) due to the combination of vertical load and lateral creep
forces since contact is made far to the field side
2-42
This problem is exacerbated when the field side of the wheel tread contacts the
center of the rail head, causing concavity of the head (figure 2-55).
Figure 2-55: Hollow Wheels Contacting the Concave Rail Head on the Low Rail
2-44
Figure 2-58: Worn Wheel Profile Contacting a Rail with a Higher Gage Shoulder
2-45
The transfer of contact between the wheel and elements of a switch can be
inhibited by the false flange that can be “jammed” between switch point and
stock rail. Instead of riding up to the stock rail, the false flange pushes the stock
rail laterally, as shown in figure 2-61, increasing the risk of stock rail rollover.2.28
2-46
Figure 2-63: Bogie Curving Performance with
Asymmetric Wheels
2-47
Considering the effects of wheel/rail conditions on vehicle performance and
wheel/rail interface previously discussed, the following parameters should be
carefully considered in the design of wheel profiles:
• Sufficient flange face angle to reduce the risk of wheel climb on curves
and switches2.30
• Proper tread shape to compromise between curving and lateral stability
on tangent track
• Strong, two-point contact on high rail of curves to be avoided
• Strong, single-point contact with small contact area either at the wheel
flange root/rail gage corner or at the wheel tread/rail crown to be
avoided
• High-contact stress to be avoided
• Smooth transition between contact zones
For a completely new rail system, wheel and rail profiles should be designed in
pairs to achieve optimal contact conditions.
Consider the population of existing wheel and rail profile shapes when
designing a new wheel profile for an existing rail system to improve wheel/rail
contact conditions and wheel/rail lives. A smooth transition from existing con-
tact conditions to a new level of equilibrium needs to be carefully designed. The
wheel and rail profiling equipment capacity (such as wheel truing machines and
grinders) in the system should also be considered for designing the transition.
As an example, the optimization of wheel profiles in North American freight
service has been pursued in several stages in the past decades. In 1988, a new
wheel profile called AAR-1B was developed to produce less strong, two-point
contact on both new and worn rails.2.31 It replaced the AAR-1:20 wheel profile as
the North American freight service wheel standard. As the concept of conformal
contact is being adopted by freight service, efforts to further modify the AAR-
1B wheel have been made. Transportation Technology Center, Inc. (TTCI) has
developed the SRI-1A wheel profile and it is being tested in service.2.32 The
National Research Council of Canada (NRC) also developed the NRC-ASW
wheel profile. Both profiles are aimed to produce more conformal contact with
the worn high rails on curves.2.1
2-48
metal removal if grinding is conducted on the new rail to remove the oxidation
layer. Figures 2-64 and 2-65 compare the new rail profiles before and after the
AREMA revision.2.33
2-49
has a tendency to cause hollow wheel tread. When the hollow wheels contact
slightly flattened, worn rail, a two-point contact with small contact areas is likely
to result, as illustrated in figure 2-66. This can lead to RCF on both wheels and
rails (figure 2-67).
Figure 2-66: Two Contact Bands with Both Having Small Contact Areas
2-50
2.8 System Issues
V2 = (g s/2l) Rc (2.11)
where:
g = acceleration due to gravity
Rc = curve radius
2l = distance between wheel/rail contact points (track gage + one
rail head width)
V = speed
2-51
2.8.1.2 Influence of Cant Deficiency/Excess Superelevation on Verti-
cal Loads
Curving under unbalance conditions results in a transfer of vertical load
between high and low rails. Unbalance is often expressed in terms of the
amount of superelevation required to restore vertical load balance from current
superelevation and speed conditions. Example:
Figure 2-69 shows:
• No centripetal acceleration or associated forces are shown. It is assumed
that balance between the prevailing vehicle speed and the ideal
superelevation for that speed has been attained. su is the difference between
this ideal superelevation and the prevailing superelevation in the curve
(see section 2.8.1.5 and figure 2-71 for definition of su).
• Figure 2-69 shows s = su as an excess of superelevation.
d = su h/2l
2-52
Figure 2-70: Lateral Component of Coupler Force
L= 2W su/2l (2.13)
Considering a car weight of 286,000 pounds, L = 2,403 pounds per inch of cant
unbalance (or for su = 1). This implies that if the lateral force were to be “shared”
equally between the four wheels of the truck, each wheel would experience a
lateral load of 2,403/4 = 601 pounds per inch of cant unbalance.
2-53
Figure 2-71: Lateral Forces on a Bogie/Car Due to
Superelevation Unbalance
The lateral component of the coupler force must be similarly “shared” between
wheel/rail contact points.
Explained in section 2.4, railway wheelsets produce lateral forces at the wheel/
rail interface by aligning at an angle of attack α to the track (figure 2-72a).
Similarly, wheelsets in bogies produce lateral forces at the wheel/rail interface
in reaction to lateral forces at the center plate by aligning at an angle of attack α
to the track (figure 2-72b). Consequently, bogies align in curves with increased
angles of attack under excess cant conditions and with reduced angles of attack
under conditions of cant deficiency.
2-54
Generally, because of the geometry of the curve and of the bogie, the angle of
attack of the lead wheelset is greater than that of the trail wheelset with the angle
of attack of the trail wheelset being approximately zero. Consequently, most of
the lateral force to counter cant unbalance is generated by the lead wheelset.
High rail flange contact creates flange forces on the lead wheelset that act in a
similar sense to forces due to curving with excess cant (figure 2-72c). This can
increase the lateral forces on the lead wheelset when curving with flange contact
under conditions of excess cant.
2-55
This would reduce the propensity for flange climb and suggests that high rail
wear and flange wear would not be increased and may be decreased.
Superelevation deficiency is generally limited to 76 mm (3 in.) in North
America with some exceptions.
Curving with excess superelevation increases the vertical and lateral loads
on lead axle, low rail contact. This is understood to be a leading cause for
high-impact wheels, particularly in North America and is also understood to be
a driver of low rail head damage through rolling contact fatigue (RCF), wear,
material flow, and crack formation.
Excess superelevation is generally limited to a static wheel load loss of 40% for
a wagon (car) at zero speed in a curve.
Given the foregoing, a design can often be obtained for balance for the
preponderance of traffic at balance speed. Figure 2-74 shows a graph of balance
speed vs. superelevation with superelevation design options to remain within
a 76 mm (3 in.) superelevation deficiency limit. It references figure 2-73 and
suggests that zero superelevation would provide balance for 83% of traffic with
a speed limit of 50 kph (31 mph) [76 mm (3 in.) superelevation deficiency];
however, if a 25.4 mm (1 in.) superelevation were provided, the speed limit would
be 56 kph (35 mph). The current design is for 89 mm (3.5 in.) superelevation
providing a maximum speed in excess of 72 kph (45 mph) and catering to 8%
of the lightly loaded traffic, but in the process introducing unnecessarily high
vertical and lateral loads on the low rail.
To reduce the effect of traction forces on vertical and lateral loads on the
contact patch, distributed power (DP) is recommended to limit in-train forces
on grades.
2-56
For the front wheelset, there is an angle of attack between the wheel plate and
the tangential line of the outer rail in the curve at the wheelset location. To a
certain extent, the size of the attack angle influences the wheel/rail lateral
dynamic interaction. A large angle of attack directly threatens the running
safety. Figure 2-76 shows the angles of attack on two curves with different radii.
Note that the angle of attack α1 in the curve with small radius is larger than that
with the big radius.
2-57
Figure 2-77: Lateral Wheelset Displacement vs. Curve Radius
2-58
• At welded rail joints, the wheel/rail impact loads occur constantly with a
high frequency force P1 and a low frequency force P2.2.41 Low vertical
stiffness of the fastener is helpful to alleviate the impact load
(figure 2-80).2.42
2-59
2.8.5.1 Asymmetries of Wheel/Rail Contacts and Forces on Curved
Track
On curved track, the wheelset usually has a lateral displacement toward the
outer rail, especially for the front wheelset, which leads to asymmetries of wheel/
rail contact. In this case, the contact parameters of the outer and inner wheels,
such as the contact positions, rolling radius, and contact angle are different from
those of the straight track case (figure 2-81).
Figure 2-82: Wheel/Rail Contact Positions of Chinese LM Wheel Profile with Stan-
dard Rail Profile
2-60
(a) Outer rail (b) Inner rail
Figure 2-83: Asymmetric Profiles of Outer and Inner Rails in a Sharp Curve
Based on wheel/rail contact theory, the asymmetric wheel/rail contact geom-
etries will obviously change the force conditions of two wheels fixed on a single
axle. Actually, the forces acting on the vehicle in curves are extremely complex,
such as the centrifugal force, wind force, wheel/rail normal force, wheel/rail
tangential force (including the longitudinal and lateral creep forces), and gravity
[figure 2-84(a)]. The primary forces ultimately on the rail can be projected to
be a wheel/rail vertical force and a lateral force [figure 2-84(b)]. They are two
important indicators for evaluating the dynamic behavior of the vehicle and the
track. The magnitudes of these two forces are significantly influenced by many
factors such as the wheel/rail contact condition, vehicle running speed, and
external forces.
In general, if the vehicle in a curve operates with a deficient superelevation, the
car body rolls and moves laterally toward the outer rail, and vehicle load transfer
occurs. In this case, the outer rail will bear a larger vertical force than the in-
ner rail. Conversely, a surplus superelevation generally leads to a vertical load
decrease on the outer rail and a load increase on the inner rail. But for the wheel/
rail lateral force, the asymmetries are closely related to the steering ability of
the wheelset. Basically, on the same curved track, the wheel/rail lateral force
of the front wheelset in the wagon bogie increases as the running speed rises.
Under certain conditions of deficient superelevation and wagon parameters, the
rear wheelset of a wagon bogie has a tendency to approach the outer rail, which
could share and reduce the lateral load on the front wheelset. In other words, the
deficient superelevation has a positive influence on the curving performance of
the wagon bogie in this case.
2-61
Figure 2-84: Schematic of Forces Acting on a Vehicle
in a Curve
The asymmetries of the wheel/rail forces on the curved track greatly depend
on the track structure parameters, vehicle performance, and wheel/rail con-
tact status. It should be analyzed under the actual track and vehicle operating
conditions.
2-62
The rail rotational motion is determined by the resultant moment M, calculated
by:
2-63
3 N 2.8.5.3 The Influence
P₀,max = 32 -of
N Track Gage
2 Ae √ 1 - ( aξ )² - ( ηb )² , Ae
A lateral clearance is kept
Δ tan(δ) - μ between the rail gage corner and wheel flange
(figure
0.3 ∙ p₀ at z = 0.78a.
2-88). When (
(32) =a max
wheelset
1 + μ *
)
tan(δ)rolls on a straight track, it will move laterally
by a distance δ to “find” equal rolling diameters. If the gage clearance Δmax is
insufficient to accommodate the distance δ, flange contact will occur, resulting
in flange and gage corner wear. The lr0
value of Δvmax and the track gage will change
Rw δ when ythe l
r0rails and λ
wheels = 2πare worn f = 0the gage is widened. From the
or after
= γ
R w - RR perspectivec of wheel/rail contact geometry,λ their influence on the wheel/
R γ
rail interaction can be illustrated qualitatively by studying the conicity
characteristic. The effective conicity γ of a wheelset is defined as the wheel
Effective Conicity (λ) = RRD
[
Δr = r0 rolling
]
R + l - 1 radius differential (RRD) divided by twice the lateral motion caused by
2y
R - l radii difference. This relationship is given in equation 2.15. In most cases, the
the
conicity increases with increasing RRD.
y R R δ2l λ π w e
R
2-64
Figure 2-89: Wheel/Rail Contact with Different Track Gages
From the aspect of geometric curving of freight wagons, gage clearance must
be maintained so that the wheelset can pass through a curve smoothly. The
influence of gage clearance on the running resistance, energy loss, and wear of
wheels and rails, etc., should always be considered.
2-65
2-66
References
2.1. Guidelines to Best Practices for Heavy Haul Railway Operations: Wheel and Rail
Interface Issues, First Edition, International Heavy Haul Association, Virginia
Beach, VA, 2001.
2.2. Tournay, H., “Wheel Rail Interaction from a Track and Vehicle Design
Perspective,” Procedings of Conference on Wheel/Rail Interface, IHHA, Moscow,
June 14-17, 1999.
2.3. Redtenbacher, F.J., “Die Gesetze des Locomotiv-Baues,” Verlag von Friedrich
Bassermann, Mannheim, p. 22, 1855.
2.4. Dendy Marshall, C.F., “A History of British Railways down to the Year 1830,”
Oxford University Press, London, pp. 147-148, 1938.
2.5. Klingel, J., “Über den Lauf von Eisenbahnwagen auf Gerader Bahn,” Organ
Fortschritte des Eisenbahnwesens, Neue Folge 20, pp. 113-123, 1883.
2.6. Andersson, E., Berg, M., and Stichel, S., “Rail Vehicle Dynamics,” KTH Railway
Group Royal Institute of Technology, Stockholm, Sweden, 2013.
2.7. Knothe, K. and Stichel, S., Schienenfahrzeugdynamik, Springer-Verlag Berlin
Heidelberg, 2003.
2.8. Scheffel, H., Tournay, H.M., and Fröhling, R.D., “The Evolution of the Three-Piece
Freight Car Bogie to Meet Changing Demands in Heavy Haul Railroads in South
Africa,” Proceedings of the Conference on Freight Car Trucks/Bogies organized by the
International Heavy Haul Association, Montreal, Canada, June 1996.
2.9. Wickens, A.H., “The Dynamic Stability of Railway Vehicle Wheelsets and Bogies
having Profiled Wheels,” International Journal of Solids and Structures, vol. 1,
pp. 319-341, 1965.
2.10. Joly, R., “Study of the Transverse Stability of a Railway Vehicle Running at High
Speed,” Rail International, vol. 3, no. 2, pp. 83-117, 1972.
2.11. Scheffel, H., “The Influence of the Suspension on the Hunting Stability of Railway
Vehicles,” Rail International, pp. 662-696, August 1979.
2.12. Scheffel, H., “Die Konstruktion der Kreuzanker-Drehgestelle der Südafrikanischen
Eisenbahnen,” ZEV-Glasers Annalen 110, Nr. 6/7, Juni/Juli 1986.
2.13. Scheffel, H. and Tournay, H.M., “An Analysis of Steady State Deviations of
Wheelsets from the Pure Rolling Position Caused by Curved Track and Bogie
Inaccuracies,” Proceedings of the 8th IAVSD Symposium, MIT, Cambridge, 1983.
2.14. Scheffel, H., Fröhling, R.D., and Heyns, P.S., “Curving and Stability Analysis of Self
Steering Bogies having a Variable Yaw Constraint,” Proceedings of the 13th IAVSD
Symposium, Chengdu, People’s Republic of China, 1993, Supplement to Vehicle
System Dynamics, volume 23, pp. 425-436, 1994.
2.15. Scheffel, H., “A New Design Approach for Railway Vehicle Suspension,” Rail
International, pp. 638-651, October 1974.
2.16. Scheffel, H., “Self-Steering Wheelset Will Reduce Wear and Permit Higher Speeds,”
Railway Gazette International, pp. 453-456, December 1976.
2.17. Scheffel, H. and Tournay, H.M., “The Development of an Optimal Profile for
Self-Steering Trucks under Heavy Axle Load Conditions,” Rail Transport Division
of the American Society of Mechanical Engineers (ASME), Paper 80-WA/RT-5,
Winter Annual Meeting, Chicago, Illinois, U.S., pp. 1-12, November 1980.
2-67
2.18. Stanley, I.N., Philadelphia, PA, “Improvement in Railroad Cars for Turning
Strong Curves, and for Changing from One Track to Another Without the Use of
Switches,” patented March 12, 1840.
2.19. Smith, R.E. and Anderson, R.J., “Characteristics of Steered Railway Trucks,” IAVSD
Journal, vol. 17, pp. 1-36, 1988.
2.20. Anderson, R.J. and Fortin, C., “Low Conicity Instabilities in Forced-Steering
Railway Vehicles,” Proceedings of the 10th IAVSD Symposium, Prague,
August 24-28, 1987.
2.21. Smith, R.E., “Steering Rail Vehicle Axles – A Historical Review,” Proceedings of the
Canadian Engineering Centennial Convention, Montreal, May 1987.
2.22. Tournay, H., “Rail/Wheel Interaction from a Track and Vehicle Design
Perspective,” IHHA STS Conference, Moscow, Russia, June 14-17, 1999.
2.23. Elkins, J.A. and Weinstock, R., “The Effect of Two Point Contact on the
Curving Behavior of Railroad Vehicles,” AMSE 82-WA/DSC-13, American Society of
Mechanical Engineers, New York, NY, 1982.
2.24. Mace, S. et al., “Effects of Wheel-Rail Contact Geometry on Wheelset Steering
Forces,” Wear, vol. 191, issues 1-2, pp. 204-209, 1996.
2.25. Wu, H. and Kerchof, B., “Management of Wheel/Rail Interface to Prevent Rail
Rollover Derailments,” 2013 IHHA conference, to be published in JRRT.
2.26. Wu, H., Woody, S., and Blank, B., “Optimization of Rail Grinding on a North
American Heavy Haul Railroad Line,” Proceedings of 2005 International Heavy Haul
Association Conference.
2.27. Sawley, K. and Wu, H., “The Formation of Hollow-Worn Wheels and their Effect on
Wheel/Rail Interaction,” Wear, 258 (7-8), pp. 1179-1186.
2.28. Tournay, H., Wu, H., and Guins, T., “The Influence of Hollow-Worn Wheels on
the Incidence and Costs of Derailments,” Research Report R-975, Association of
American Railroads, Transportation Technology Center, Inc., Pueblo, CO, U.S.,
February 2004.
2.29. Wu, H. and Madrill, B., “Effect and Formation of Asymmetrically Worn Wheels,”
Technology Digest TD-08-021, Association of American Railroads, Transportation
Technology Center, Inc., Pueblo, CO, U.S., 2008.
2.30. Blader, F.B., “A Review of Literature and Methodologies in the Study of
Derailments Caused by Excessive Forces at the Wheel/Rail Interface,” Research
Report R-717, Association of American Railroads, Transportation Technology
Center, Inc., Pueblo, CO, U.S., December 1990.
2.31. Leary, J., “Final Report on the Development of an Alternative AAR Interchange
Wheel Profile,” Report No. R-706, Association of American Railroads, November
1988.
2.32. Wu, H., “Effects of Wheel and Rail Profiles on Vehicle Performance,” Proc.
International Association of Vehicle System Dynamics IAVSD Conference, Milano,
Italy, August 2005.
2.33. American Railway Engineering and Maintenance-of-Way Association Manual, 2005,
2006.
2.34. Sroba, P. and Roney, M. et al., “The Evolution of Rail Grinding on Canadian Pacific
Railway to Address Deep Seated Shell in 100% Effective Lubrication Territory,”
World Congress on Railway Research, Montreal, Quebec, Canada, 2006.
2-68
2.35. Zhai, W.M., Wang, K.Y., and Cai, C.B., “Fundamentals of Vehicle-Track Coupled
Dynamics,” Vehicle System Dynamics, 47(11), pp. 1349-1376, 2009.
2.36. Zhai, W.M., Cai, C.B., and Guo, S.Z., “Coupling Model of Vertical and Lateral
Vehicle/Track Interactions,” Vehicle System Dynamics, 26(1), pp. 61-79, 1996.
2.37. Zhai, W.M. and Wang, K.Y., “Lateral Interactions of Trains and Tracks on
Small-Radius Curves: Simulation and Experiment,” Vehicle System Dynamics,
44 (Suppl.), pp. 520-530, 2006.
2.38. Zhai, W.M. and Wang, K.Y., “Lateral Hunting Stability of Railway Vehicles Running
on Elastic Track Structures,” Journal of Computational and Nonlinear Dynamics,
ASME, 5(4), pp. 041009-1-9, 2010.
2.39. Andersson, E., Berg, M., and Stichel, S., “Rail Vehicle Dynamics,” KTH Railway
Group Royal Institute of Technology, Stockholm, Sweden, 2013.
2.40. Knothe, K. and Stichel, S., Schienenfahrzeugdynamik, Springer-Verlag Berlin
Heidelberg, 2003.
2.41. True, H., “On the Theory of Non-Linear Dynamics and its Applications in Vehicle
System Dynamics,” Veh. Syst. Dyn., 31(5), pp. 393-422, 1999.
2.42. Polach, O., “On Non-Linear Methods of Bogie Stability Assessment using
Computer Simulations,” Proceedings of the Institution of Mechanical Engineers, Part
F: Journal of Rail and Rapid Transit, 220, pp. 13-27, 2006.
2.43. Jenkins, H.H. et al., “The Effect of Track and Vehicle Parameters on Wheel/Rail
Vertical Dynamic Forces,” Railway Engineering Journal, 3(1): 2-16, 1974.
2-69
2-70
Contributors to Chapter 2
Harr y Tournay
Mr. Tournay is Senior Scientist at Transportation
Technology Center, Inc., and for the past 12 years has
conducted mechanical and vehicle/track research under
the Association of American Railroads strategic research
program. Projects include the development of an improved
freight bogie, investigating the root causes of (rail and
wheel) rolling contact fatigue, investigating the effects of
superelevation on freight vehicle/track interaction, brake
technology, and wayside detection.
Robert Fröhling
Dr. Fröhling has more than 30 years of railway experience with extensive
experience in rail vehicle system dynamics, vehicle/track interaction, wheel/rail
interaction, bogie technology, and structural mechanics. He has published and/
or presented 47 international papers on these subjects. He is Principal Engineer
within the Technology Management Department of Transnet Freight Rail and
he has responsibility for the following core Mechanical Railway Technologies:
Railway Vehicle System Dynamics, Wheel/Rail Interaction, Bogie Technology,
Structural Mechanics, as well as Locomotive and Wagon Mechanical Design
Integrity.
2-71
Roy Smith
Mr. Smith, P.Eng., M.Eng., B.Sc. has been the president of RESCO Engineer-
ing since he formed the company in 1989. Prior to that, between 1975 and 1989,
he worked on rail car truck design at UTDC (now part of Bombardier). He
graduated from the University of Birmingham (UK) in 1963 and from the
University of Toronto in 1970. The author of many technical papers on the
subject of truck dynamics and curving, Mr. Smith also has over twenty patents
to his name. He continues to be active in the development of technology for
the improvement of railway economics and safety. He was born in England and
emigrated to Canada in 1965, where he has lived ever since.
Sebastian Stichel
Prof. Stichel holds a Ph.D. (1996) in Vehicle Dynamics from TU Berlin.
He has been a professor at the Royal Institute of Technology in Stockholm
since 2010. He is director of the KTH Railway Group, vice chairman of the
Department of Aeronautical and Vehicle Engineering, and director of the mas-
ter’s program in Vehicle Engineering at KTH. From 2000 to 2010, Dr. Stichel
was employed at Bombardier Transportation in Sweden where he headed a
Vehicle Dynamics department with employees in Sweden, Germany, UK, and
France since 2003. His primary research interest is in dynamic vehicle/track
interaction using multi-body simulation including improved ride comfort and
reduced wheel and track damage.
Huimin Wu
Dr. Wu is a Principal Investigator at Transportation Technology Center, Inc.,
Pueblo, Colorado, U.S. She holds a Ph.D. degree in Mechanical Engineering
from the Illinois Institute of Technology, Chicago, U.S. Dr. Wu has more than
20 years of experience in railroad research areas of vehicle dynamics, wheel/
rail interaction, wheel/rail profile design, rail grinding, automated wheel/rail
contact inspection systems, and rail wear and RCF management model.
Wanming Zhai
Prof. Zhai is a Professor at Southwest Jiaotong University, a member of the
Chinese Academy of Sciences and the Editor-in-Chief of International
Journal of Rail Transportation. He has developed a model of vehicle-track coupled
dynamics and a method to analyze and evaluate high-speed train-track-bridge
dynamic interaction for Chinese high-speed operations. He graduated from
Southwest Jiaotong University in 1985, receiving a Ph.D. in Railway Vehicle
Engineering in 1992.
2-72