Upload
truonglien
View
229
Download
4
Embed Size (px)
Citation preview
On the Fundamentals of Track Lateral Resistance
ABSTRACT
A key aspect of track lateral stability assurance is the provision of adequate lateral resistance to
the rail-tie structure by the ballast. When lateral resistance is reduced, such as when ballast is
worked, the track cannot only lose alignment and surface, but can become buckling prone in the
presence of high thermal forces. Current methods to restore ballast strength after track work is
through mechanical compaction through dynamic track stabilization (DTS) or through traffic (train
load/tonnage) induced consolidation. Research has confirmed the effectiveness of DTS, however,
due to the complex nature of ballast compaction mechanics under train loads, a reliable
quantification of the effectiveness of train/tonnage based consolidation is still missing.
This paper presents a review of the fundamentals of track lateral resistance, its measurement and
its key influencing parameters, provides a test/data summary of US measurements to date,
present analysis results on influence on track stability, and offers insights on research needs for
further research studies for track lateral resistance improvements.
Key Words: track lateral resistance, ballast maintenance, track buckling, track shift, lateral
resistance measurement, STPT, track stabilization
Word Count: 4253
Andrew Kish, Ph.D. President - Kandrew Inc. Consulting Services
Peabody, MA, USA Phone: 978-535-3342
Email: [email protected]
© 2011 AREMA ®
1.0 INTRODUCTION
Track resistance is one of the most important parameters influencing track performance and
safety. Through its three in-plane components of lateral, longitudinal, and torsional, and one out-
of-plane of vertical, it influences most track lateral stability aspects including geometry retention
and buckling prevention. The three in-plane components are schematically shown in Figure 1
with their typical “spring” analogy representations.
Figure 1- Track Lateral Resistance Spring Analogy Longitudinal resistance provided to the rail/tie structure by fasteners and anchors is important to
prevent rail running (and thus limit rail neutral temperature variation), and to limit tensile break
gap sizes in the event of rail breaks. Torsional resistance provided by the spiked tie plates,
anchors and fasteners against rail rotation (in-plane-bending) also offers “rigidity” to the track
structure, especially against buckling.
Lateral resistance is the reaction offered by the ballast against lateral movement, often referred to
as the “lateral strength”. It is a tie-ballast interaction parameter which is influenced by several
factors such as ballast section, condition, consolidation, maintenance, tie type and condition, and
train loads. As such, track lateral resistance becomes a fundamental but highly variable
parameter in track lateral stability assurance, and is the focus of this paper.
P
P
P
P
Lateral spring
Torsional spring Longitudinal spring
Track Resistance Parameters
§ Lateral resistance (ties against ballast)
§ Longitudinal resistance (rails through fastenersand ties through ballast)
§ Torsional resistance (fasteners resisting rail rotation)
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
Lateral springLateral spring
Torsional springTorsional spring Longitudinal springLongitudinal spring
Track Resistance Parameters
§ Lateral resistance (ties against ballast)
§ Longitudinal resistance (rails through fastenersand ties through ballast)
§ Torsional resistance (fasteners resisting rail rotation)
P
P
P
P
Lateral spring
Torsional spring Longitudinal spring
Track Resistance Parameters
§ Lateral resistance (ties against ballast)
§ Longitudinal resistance (rails through fastenersand ties through ballast)
§ Torsional resistance (fasteners resisting rail rotation)
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
Lateral springLateral spring
Torsional springTorsional spring Longitudinal springLongitudinal spring
Track Resistance Parameters
§ Lateral resistance (ties against ballast)
§ Longitudinal resistance (rails through fastenersand ties through ballast)
§ Torsional resistance (fasteners resisting rail rotation)
© 2011 AREMA ®
The paper will present a review of the fundamentals of track lateral resistance, its measurement
and parametric influences, provide a test/data summary of US measurements to date, present
analysis results on influences on track stability, and offer insights on research needs for further
quantification studies for track stability safety and maintenance improvements.
2.0 LATERAL RESISTANCE FUNDAMENTALS
2.1 Basic Response Characteristics
As indicated in Figure 1, lateral resistance can be envisioned as a “spring” load-deflection
response characteristic representing tie-ballast interaction. As over 30 years research and tests
have confirmed, it is a load versus deflection non-linear “spring” response as illustrated in Figure
2, including a simplified analytic representation.
Figure 2 – Lateral Resistance Load Deflection Characteristics
Key features include an initial linear stiffness, a peak (break-away) value, and a decreasing
(drooping) portion to a constant “limit” value. Representations of typical “strong”, “average”,
“weak” behaviors for concrete ties are also shown. Key parameters/conditions influencing the
behavior include: tie type, weight, shape and spacing; ballast type/condition (fouled, wet, frozen,
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
Ap
pli
ed
Lo
ad
(lb
s)
Limit Resistance
weak
Initial Stiffness average
Peak Resistance
strong
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
Ap
pli
ed
Lo
ad
(lb
s)
Limit Resistance
weak
Initial Stiffness average
Initial Stiffness average
Peak Resistance
strong
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
Ap
pli
ed
Lo
ad
(lb
s)
Limit Resistance
Initial Stiffness
Peak Resistance
Idealization
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
Ap
pli
ed
Lo
ad
(lb
s)
Limit Resistance
Initial Stiffness
Peak Resistance
Idealization
© 2011 AREMA ®
etc.), shoulder width, crib content; maintenance, degree of consolidation, and train loads. Typical
values and ranges are shown in Figure 3.
Figure 3 – Typical Peak Lateral Resistance Values
Lateral resistance has three contributing components of bottom friction (Fb), side friction (Fs), and
end or shoulder restraint (Fe), and their approximate contributions are as indicated in Figure 4. As
research has shown, the bottom friction component is most influenced by tie type and weight, the
side friction by crib content, and the end restraint by ballast shoulder geometry. Consolidation and
compaction influence all of the three components to different degrees.
Figure 4 – Lateral Resistance Components
2.2 Lateral Resistance Measurement
Over the years several measurement techniques have been developed, including:
• single tie push test (STPT)
Fside Fbottom Fend
F F = Fs + Fb + Fe
20-25%35-40%30-35%Fside Fbottom Fend
F
Fside Fbottom Fend
F F = Fs + Fb + Fe
20-25%35-40%30-35%
F = Fs + Fb + FeF = Fs + Fb + FeF = Fs + Fb + Fe
20-25%35-40%30-35%
Typical concrete tie peak values (static)
1800 2000 2200 2400 2600 2800 3000 3200 3400 3600
Weak Marginal* Average Strong
7.5 10.0 12.5 15.0
lbs/tie
kN/tie
Note: for timber ties subtract 500 lbs/tie * Typical consolidation range
Typical concrete tie peak values (static)
1800 2000 2200 2400 2600 2800 3000 3200 3400 3600
Weak Marginal* Average StrongStrong
7.5 10.0 12.5 15.0
lbs/tie
kN/tie
Note: for timber ties subtract 500 lbs/tie * Typical consolidation range
© 2011 AREMA ®
• discrete cut panel pull test
• continuous track panel pull test (TLPT)
• continuous dynamic measurement (Plasser-DGS)
• analytic empirical model
The most advantageous and frequently used technique is the STPT which mobilizes a single tie
in the ballast and measures the force versus displacement behavior as shown in Figure 1. This
provides the non-linear “spring” type characteristic required in track stability analyses. Figure 5
shows the STPT device developed by the Volpe Center in the US, and for detailed descriptions
refer to [1, 2].
The discrete cut panel pull test requires cutting rails and is highly destructive, while the TLPT
measures an entire continuous panel’s load-deflection response which includes the rail flexural
rigidity, the rail thermal force, and the non-uniform resistance offered by several ties. Thus from
TLPT data the single tie resistance is not calculable. The recently developed continuous
measurement technique through Plasser’s Dynamic Track Stabilizer (DTS, DGS) is an indirect
measurement relating the DTS energy output to the “frictional power” to move ballast in the track
grid. For a more detailed description on this measurement and tests refer to [3, 4]. The analytic
empirical model is lateral resistance estimator based on empirical equations developed through
over 500 STPT measurements in the US which evaluated the influences of shoulder width, crib
content, and consolidation levels on lateral resistance for both wood and concrete ties in granite
ballast. The model is available in the CWR-INDY module of the CWR-SAFE program which is
discussed in 4.2.
© 2011 AREMA ®
Figure 5 – Single Tie Push Test (STPT) Measurement
2.3 Concept of “Dynamic” Lateral Resistance and Tie/Ballast Friction Coefficient Train loads and dynamics play a key role in track lateral stability behavior due to their impact on
lateral resistance. Dynamic lateral resistance refers to the increase/decrease in resistance due to
train loads as illustrated in Figure 6.
Figure 6 – Dynamic Resistance Concept
The impact of dynamic uplift is that a percentage of the tie bottom component can be lost,
whereas under load the resistance is increased. This increase has been shown to be related to a
tie-ballast friction coefficient parameter, µ, as indicated in Figure 7, and as discussed in [5, 6].
Based on resistance measurements on vertically loaded ties, this parameter has been shown to
Dynamic Uplift
Reduced ResistanceIncreased Resistance Increased Resistance
Dynamic Uplift
Reduced ResistanceIncreased Resistance Increased Resistance
Dynamic Uplift
Reduced ResistanceIncreased Resistance Increased Resistance
© 2011 AREMA ®
be vertical load and tie type/condition dependent. The parameter µ plays an important role in both
buckling and track shift analyses as discussed later.
Figure 7 – Dynamic Resistance Friction Coefficient Behavior
Another aspect of dynamic lateral resistance is the influence of train induced vibration. Although
the effect is difficult to measure and quantify, tests by British Rail under the ERRI/D202 work
indicated a 25-50% reduction in lateral resistance depending on the acceleration and vibration
frequency levels [7]. Such reductions are currently not included in track stability analyses, and
further quantification would be most useful for US heavy tonnage tracks/equipment.
3.0 LATERAL RESISTANCE AND BALLAST CONSOLIDATION
When lateral resistance is reduced, such as when ballast is worked, the track can not only lose
alignment and surface, but can become buckling prone in the presence of high thermal forces.
Current methods to restore ballast strength after track work are through dynamic track
stabilization (DTS), or through traffic (train-load) induced consolidation. Research has shown both
methods to be effective to various degrees, however due to the complex nature of ballast
Fdyn = Fstat + µRV
RV
Fdyn
RV
Friction coefficient, µ, is a tie bottom roughness index
Fdyn = Fstat + µRV
RV
Fdyn
RV
Friction coefficient, µ, is a tie bottom roughness index
Fdyn = Fstat + µRV
RV
Fdyn
RV
Friction coefficient, µ, is a tie bottom roughness index
© 2011 AREMA ®
compaction mechanics under train loads and to its many influencing components, a reliable
quantification of MGT effectiveness is still missing. Such information is a prerequisite for the
effective placement and removal of slow orders. Key points addressing ballast lateral resistance
restoration include:
• Surfacing and tamping can reduce track lateral resistance (TLR) by 40-70%. The actual
amount depends on ballast type, ballast section/condition, wood vs. concrete ties, tie
design and condition, etc, i.e. on the initial conditions before surfacing. It also depends on
the nature of the ballast disturbance/work. As an example in the case of surfacing, the
amount of lift is considered to have a large influence. Therefore the key issues are: (1)
how much is the reduced resistance from which resistance restoration is required, (2)
how fast can it be regenerated to “safe” levels, and (3) what are the safe levels?
• DTS has been shown to regenerate 30-50% of resistance (from the reduced value)
based on European data. Limited US data is in the similar range, as discussed in 3.2.
• DTS tonnage equivalent (European tests – UIC Leaflet #720) is on the order of 0.1 MGT
(million gross tons) of traffic. However this equivalence has not been conclusively
demonstrated for US tracks/conditions as indicated in 3.2.
3.1 Ballast Consolidation Mechanism
For DTS the basic mechanism for resistance restoration is a “reintensification of the tie-bottom
contact”. The DTS horizontal vibratory input under a specified vertical load causes a dynamic re-
arrangement of the ballast particles and an increases tie-bottom contact area [3]. Such tie–bottom
contact restoration effectively increases the tie/ballast friction coefficient represented in Figure 7,
as well as assures the 35-40% tie bottom resistance contribution as referred to in Figure 4.
© 2011 AREMA ®
Train traffic induced (MGT) consolidation is however a different mechanism altogether. The wheel
induced dynamic uplift versus downward pressure coupled with vertical vibration input presents a
different compaction mechanics mode, and remains a minimally researched topic to date. Hence
the questions to resolve include: (1) how traffic tonnage consolidates (by what mechanism), and
(2) how effectively (at what tonnage rate)?
3.2 US Data on Maintenance, DTS and MGT Influences on Lateral Resistance
Over the past 30 years the US railroads, the US DOT’s Volpe Center, and the Association of
American Railroads have conducted numerous STPT tests to evaluate lateral resistance
behavior. Aside from the fundamental mechanistic aspects, a large focus was placed on
evaluating the influences of maintenance and subsequent restoration of lateral resistance through
DTS and tonnage. The following is a brief synopsis of some applicable results:
1. Chessie and Amtrak Tests – concrete ties (1985, Sluz, [8])
a. Chessie: 9% recovery after 0.076MGT
b. Amtrak: 11% recovery after 0.073MGT
2. AAR/TTC Tests – wood ties (1990 – Trevizo, [9])
a. Tangent: 17% recovery after 0.1MGT; 32% after 1MGT
b. 5° Curve: 9% recovery after 0.1MGT; 21% after 1MGT
3. Volpe/FRA Tests at TTC - wood and concrete (1987-1990; Kish, et al, [10])
a. Wood - tangent: 26% recovery after 0.1MGT
b. Concrete – 5° Curve: 52 % reduction due to tamping
c. Concrete – 5° Curve: 22% recovery after 0.1MGT
© 2011 AREMA ®
4. Volpe/Union Pacific Tests – concrete (2000-Sluz, [11])
a. Concrete (new-scalloped) -17% recovery after 0.35MGT
b. DTS increase - 33%
5. Volpe/Amtrak/FRA Tests – concrete (2001- Kish, et al, [12, 13])
a. 43% reduction due to surfacing with ½ inch lift
b. DTS increase: 31%
6. UP/Foster-Miller Tests – wood/concrete/old/new/DTS (2001-Samavedam [14])
a. 39-70% reduction due to tamping/surfacing
b. 0.1MGT had negligible influence on wood after tamping; 0.2MGT was 28%
c. DTS increase on concrete: 22%
d. After DTS, traffic decreased lateral resistance initially, then it increased
e. Heavy rain/wet ballast: 20% decrease in lateral resistance on wood
Some key features of items 3 and 5 are illustrated in Figures 8 and 9. As the above summary
indicates there is a large variability on traffic/MGT influence on track lateral resistance recovery
after maintenance, and it appears that a larger than 0.1 MGTs may be required for a DTS
equivalent of 30-50% lateral resistance restoration. Thus while the DTS benefits are relatively
reliable and consistent, there are several traffic/tonnage based consolidation issues/questions not
resolved to date. These include: (1) is there a train speed influence on consolidation, (2) is there
an axle load influence, (3) is there a curvature influence, (4) is there a track material influence i.e.
ballast type, tie type, etc., and (5) is there an “initial condition” influence i.e. for example a skin lift
might produce a 30% decrease in resistance, whereas a major surfacing might be a 60%
reduction, both requiring different recovery rates. Tests are needed to better evaluate and
quantify these parameters/conditions. The key railroad impact issue is that there is an urgent
need for such data to manage speed restrictions after maintenance quickly an effectively without
compromising safety.
© 2011 AREMA ®
Figure 8 – Concrete Tie MGT Consolidation Tests at FAST (Ref. 10)
Figure 9 – Concrete Tie Track Stabilization Tests on Amtrak/New Carrollton (Ref. 12)
0
5
10
15
20Min: 1700Max: 2200Mean: 1926Std Dev: 136Count: 42
Post-Maintenance(surfacing)
0
5
10
15
20
Min: 2200Max: 2900Mean: 2520Std Dev: 240Count: 35
Post-DTS
0
5
10
15
20
Min: 1900Max: 2500Mean: 2150Std Dev: 163Count: 10
No DTS- 3,360 Tons Traffic
Nu
mb
er
of
ST
PT
s
Pre-Maintenance
0
5
10
15Min: 2700Max: 4000Mean: 3393Std Dev: 330Count: 37
1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800 4000
Peak lateral resistance (lbs/tie)
0
5
10
15
20
0
5
10
15
20Min: 1700Max: 2200Mean: 1926Std Dev: 136Count: 42
Post-Maintenance(surfacing)
0
5
10
15
20
0
5
10
15
20
Min: 2200Max: 2900Mean: 2520Std Dev: 240Count: 35
Post-DTS
0
5
10
15
20
Min: 1900Max: 2500Mean: 2150Std Dev: 163Count: 10
No DTS- 3,360 Tons Traffic
Nu
mb
er
of
ST
PT
s
Pre-Maintenance
0
5
10
15Min: 2700Max: 4000Mean: 3393Std Dev: 330Count: 37
1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800 4000
Peak lateral resistance (lbs/tie)
Consolidation (MGT)
0 2 4 6 8 10
Pe
ak la
tera
l resis
tance
incre
ase
(lb
s/tie
)
0
200
400
600
800
1000
1200
1400
Fitted equation
Data
Initial tamped@ 2300lbs/tie
Note: At each MGT increment an average of 15 STPTs is shown
22% increase due to 0.1MGT
52% loss due to tamping @ 9.7MGTs
Peak
late
ral r
esi
stance incr
ease a
fter
tam
pin
g (
lbs/
tie)
Consolidation (MGT)
0 2 4 6 8 10
Pe
ak la
tera
l resis
tance
incre
ase
(lb
s/tie
)
0
200
400
600
800
1000
1200
1400
Fitted equation
Data
Initial tamped@ 2300lbs/tie
Note: At each MGT increment an average of 15 STPTs is shown
22% increase due to 0.1MGT
52% loss due to tamping @ 9.7MGTs
Consolidation (MGT)
0 2 4 6 8 10
Pe
ak la
tera
l resis
tance
incre
ase
(lb
s/tie
)
0
200
400
600
800
1000
1200
1400
Fitted equation
Data
Initial tamped@ 2300lbs/tie
Note: At each MGT increment an average of 15 STPTs is shown
22% increase due to 0.1MGT
52% loss due to tamping @ 9.7MGTs
Peak
late
ral r
esi
stance incr
ease a
fter
tam
pin
g (
lbs/
tie)
© 2011 AREMA ®
4.0 LATERAL RESISTANCE INFLUENCE ON TRACK STABILITY
Track lateral stability assurance requires effective lateral strength management. The track
stability mechanism features are outlined in Table 1, where track lateral resistance is a contributor
to all three events of the lateral stability mechanism.
Table 1 – Track Lateral Stability Mechanism
The mechanistic difference between track shift and track buckling is in the principal load
mechanism producing the event, namely the many cycles of L/V’s for track shift versus the high
longitudinal forces for track buckling. Also, track buckling typically requires the presence in initial
line defects which are usually generated by track shift. Both track shift and track buckling events
are highly impacted by lateral resistance, but in different aspects as discussed below.
1) High L/V’s2) Reduced local lateral resistance3) Initial imperfections (welds), construction
anomalies, and install errors
1 Formation of initialtrack misalignments
1) Increase in L/V, and high longitudinal forces2) Reduced lateral resistance at line defects3) Track “dynamic uplift”4) Many cycles of L/V’s
2Growth of
misalignments(Track Shift)
1) High longitudinal force2) Reduced TN (stress-free temperature)3) Weakened lateral resistance4) Train loads and dynamics5) Misalignments generated by track shift
3 Buckling
Event Major causal factorsStage
Track Lateral Stability Mechanism
1) High L/V’s2) Reduced local lateral resistance3) Initial imperfections (welds), construction
anomalies, and install errors
1 Formation of initialtrack misalignments
1) Increase in L/V, and high longitudinal forces2) Reduced lateral resistance at line defects3) Track “dynamic uplift”4) Many cycles of L/V’s
2Growth of
misalignments(Track Shift)
1) High longitudinal force2) Reduced TN (stress-free temperature)3) Weakened lateral resistance4) Train loads and dynamics5) Misalignments generated by track shift
3 Buckling
Event Major causal factorsStage
Track Lateral Stability Mechanism
© 2011 AREMA ®
4.1 Lateral Resistance Influence on Track Shift and Lateral Alignment Retention
Track lateral shift is the formation and growth of lateral track misalignments due to high lateral to
vertical load ratios (net axle L/V’s or NAL’s) coupled with longitudinal forces. Usually many axle
passes and high L/V’s are required to produce a misalignment. Typical description of track shift is
in terms of cumulative lateral deflections versus the number of axle passes is shown in Figure 10
together with the influencing parameters, including the lateral resistance components.
Figure 10 - Track Lateral Shift Parameters and L/V Response
The lateral resistance function and its idealization for track shift analyses have to account for both
the loaded and unloaded resistance, hence the importance of the tie/ballast friction coefficient
parameter. Figure 11 depicts the lateral resistance elements for track shift. As research has
shown, the key lateral resistance parameters for the determination of cumulative deflection due to
net axle L/Vs are the slopes defining the elastic and the peak limits as indicated by Fe, we, Fp and
wp on Figure 11. The lateral resistance elastic limit (Fe and we) is most important in track shift
response because it governs the ballast load-unload hysteresis characteristics, hence the
residual deflections. The analysis to predict track lateral shift response is by the US DOT/Volpe
model called TREDA (short for Track Residual Deflection Analysis). For a detailed description of
0.0
0.1
0.2
0.3
0.4
0.5C
um
ula
tiv
e lat
era
lre
sid
ua
l de
fle
ctio
n (
in)
0 100 200 300 400 500
Number of axle passes
Low L/V
Moderate L/V
High L/V
• Rail section properties
• Thermal load
• Track type and curvature
• Initial misalignments
• Vertical foundation modulus
• Type of load and number of load cycles (axle passes)
• Tie lateral resistance: (with and without vertical load)
Ø Elastic limit: load/unloadhysteresis loop
Ø Tie ballast friction coefficient
Track Lateral Shift Response and Parameters
0.0
0.1
0.2
0.3
0.4
0.5C
um
ula
tiv
e lat
era
lre
sid
ua
l de
fle
ctio
n (
in)
0 100 200 300 400 500
Number of axle passes
Low L/V
Moderate L/V
High L/V
0.0
0.1
0.2
0.3
0.4
0.5C
um
ula
tiv
e lat
era
lre
sid
ua
l de
fle
ctio
n (
in)
0 100 200 300 400 500
Number of axle passes
Low L/V
Moderate L/VModerate L/V
High L/V
• Rail section properties
• Thermal load
• Track type and curvature
• Initial misalignments
• Vertical foundation modulus
• Type of load and number of load cycles (axle passes)
• Tie lateral resistance: (with and without vertical load)
Ø Elastic limit: load/unloadhysteresis loop
Ø Tie ballast friction coefficient
• Rail section properties
• Thermal load
• Track type and curvature
• Initial misalignments
• Vertical foundation modulus
• Type of load and number of load cycles (axle passes)
• Tie lateral resistance: (with and without vertical load)
Ø Elastic limit: load/unloadhysteresis loop
Ø Tie ballast friction coefficient
Track Lateral Shift Response and Parameters
© 2011 AREMA ®
TREDA, its validation, parametric studies, and for its applications to determine maximum
allowable L/Vs to prevent excessive residual lateral displacements refer to [5, 6, 15, 16].
Figure 11 – Lateral Resistance Elements for Track Shift Analysis
An example on the influence of lateral resistance elastic limit on track lateral deflection response
is illustrated in Figure 12. The response is highly elastic limit dependent, and the figure shows
that for the parameters chosen to represent a “weak” resistance, 2° curve, concrete tie track
under summer conditions, an L/V=0.5 (vertical axle load of 55kips and a lateral axle load of 27.5
kips) is not permissible due incurring progressive lateral deflections.
Lateral Deflection, W
Late
ral R
esis
tan
ce, F
WpWe
Fe
Fp
Slope, K2
Slope, K1
Fp (stat)
Unloaded tie
Fe
We Wp
Lateral Deflection, W
Fp (dyn)
µ • Rv
Loaded tie
Tie ballastfrictioncoefficient
Tie load
Lateral Deflection, W
Late
ral R
esis
tan
ce, F
WpWe
Fe
Fp
Slope, K2
Slope, K1
Fp (stat)
Unloaded tie
Fe
We Wp
Lateral Deflection, W
We Wp
Lateral Deflection, W
Fp (dyn)
µ • Rv
Loaded tie
Tie ballastfrictioncoefficient
Tie load
© 2011 AREMA ®
Figure 12 – Elastic Limit Influence on Track Shift Response
4.2 Lateral Resistance Influence on Track Buckling
It is well established that track lateral resistance plays a key role in track buckling. Many buckling
derailments are attributable all or in part to weakened ballast condition, hence to a reduced lateral
resistance. In order to better portray the impact of lateral resistance on track bucking, it is
important to provide a brief synopsis of track buckling fundamentals.
4.2.1 The Buckling Mechanism
Track bucking is best described through a stability diagram that portrays all the positions of
equilibrium as illustrated in Figure 13 in terms of temperature increase above neutral (TN) versus
lateral displacement.
0 10 20 30 40 50 60 70 80 900.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Resid
ual d
eflection (
in)
Number of axle passes
We1=0.005 in.
We1=0.03 in.
We1=0.05 in.
We1=0.015 in.
2° curve
concrete tie track
136RE rail
24 in. tie spacing
Kv=10000 psi
Fp=2000 lbs, wp=0.30 in.
Fe=1500 lbs, we1=varies
we2=0.1 in.
µ1=0.9, µ2=0.75, L/V=0.5
∆T=50°F, V=55 kips
constant lateral force
5000
0 10 20 30 40 50 60 70 80 900.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Resid
ual d
eflection (
in)
Number of axle passes
We1=0.005 in.We1=0.005 in.
We1=0.03 in.We1=0.03 in.
We1=0.05 in.We1=0.05 in.
We1=0.015 in.
2° curve
concrete tie track
136RE rail
24 in. tie spacing
Kv=10000 psi
Fp=2000 lbs, wp=0.30 in.
Fe=1500 lbs, we1=varies
we2=0.1 in.
µ1=0.9, µ2=0.75, L/V=0.5
∆T=50°F, V=55 kips
constant lateral force
5000
2° curve
concrete tie track
136RE rail
24 in. tie spacing
Kv=10000 psi
Fp=2000 lbs, wp=0.30 in.
Fe=1500 lbs, we1=varies
we2=0.1 in.
µ1=0.9, µ2=0.75, L/V=0.5
∆T=50°F, V=55 kips
constant lateral force
5000
© 2011 AREMA ®
Figure 13 – Track Buckling Mechanism Stability Behavior
The two equilibrium positions of interest are the peak and minimum points on the curve denoted
by Tbmax and Tbmin which define the “buckling regime” in which buckling takes place. Tbmin is
referred to as the “safe allowable temperature increase” in accordance with structural stability
criterion. Current track buckling safety criterion is in fact based on Tbmin plus a safety factor [17].
As expected, Tbmax,Tbmin and the buckling regime are highly track parameter dependent, where
one key parameter is lateral resistance.
4.2. 2 Lateral Resistance Influence on Buckling Temperatures
Over the years numerous analytic studies have been performed on evaluating lateral resistance
parametric aspects on buckling temperatures [2, 18], and here just a few salient points will be
illustrated. It is most important to recognize the importance and influence of both the peak and the
limit resistances which is illustrated in Figure 14 and described in [18]. Principally, the peak
resistance controls Tbmax, where as the limit resistance strongly influences Tbmin, the safe
temperature increase value. Hence since Tbmin determines buckling safety through Tall, the
knowledge of the full lateral characteristic is critical.
Lateral Displacement
TR
TN
Safe Allowable
Temperature Increase
(BUCKLING STRENGTH)
Buckling
regime
Tbmin
Tbmax
BUCKLE
Low Speed
High Speed
Lateral Displacement
TR
TN
Lateral Displacement
TR
TN
Lateral Displacement
TR
TN
BUCKLE
Low Speed
High Speed
Safe AllowableTemperature Increase, Tall
(BUCKLING STRENGTH)
Buckling regime
Tbmax
Tbmin
TR
TN
Lateral Displacement
TR
TN
Lateral Displacement
TR
TN
Lateral Displacement
TR
TN
TR
TN
Safe Allowable
Temperature Increase
(BUCKLING STRENGTH)
Buckling
regime
Tbmin
Tbmax
BUCKLE
Low Speed
High Speed
Lateral Displacement
TR
TN
Lateral Displacement
TR
TN
TR
TN
Lateral Displacement
TR
TN
TR
TN
Lateral Displacement
TR
TN
BUCKLE
Low Speed
High Speed
Safe AllowableTemperature Increase, Tall
(BUCKLING STRENGTH)
Buckling regime
TbmaxTbmax
Tbmin
TR
TN
© 2011 AREMA ®
Figure 14 – Peak and Limit Resistance Impacts on Buckling Temperatures Since in most practical applications it’s difficult to evaluate the large displacement limit resistance
values, correlation statistics have been developed to afford easier analytic treatments. The
resulting empirical relationships are based on the over 500 STPT measurement in the US, both at
TTC/FAST and in revenue service. Additional correlations with similar results were established by
British Rail (BR) and Deutsche Bahn (DB) [19]. The US measured correlation equations relating
FP and FL are given from [2] as:
Buckling analyses embodying the non-linear aspects of the lateral resistance and its tie/ballast
friction coefficient based dynamic effects are available through the US DOT/Volpe Center
developed CWR-SAFE model [20]. Some key lateral resistance parametric influences are
illustrated below.
0 1 2 3 4
Displacement, w (in)
0
500
1000
1500
2000
2500
3000
3500
Late
ral R
esis
tan
ce
, F
(lb
s)
Peak Resistance, FP
Limit Resistance, FL
WP WL
Lateral Displacement
TR
TN
Bucklingregime
Tall
Tbmax
Tbmin
0 1 2 3 4
Displacement, w (in)
0
500
1000
1500
2000
2500
3000
3500
Late
ral R
esis
tan
ce
, F
(lb
s)
Peak Resistance, FP
Limit Resistance, FL
WP WL
Lateral Displacement
TR
TN
Bucklingregime
Tall
Tbmax
Tbmin
Concrete
<
≥+=
1530 lbs FifF
lbs1530Fiflbs950F38.0F
PP
PP
L
<
≥+=
715 lbsFifF
715 lbsFiflbs500F3.0F
PP
PP
L
Wood Concrete
<
≥+=
1530 lbs FifF
lbs1530Fiflbs950F38.0F
PP
PP
L
<
≥+=
715 lbsFifF
715 lbsFiflbs500F3.0F
PP
PP
L
<
≥+=
715 lbsFifF
715 lbsFiflbs500F3.0F
PP
PP
L
Wood
© 2011 AREMA ®
(a) Buckling Response Behavior for “Weak”, “Average” and “Strong” Resistance Tracks
Figure 15 shows the buckling response comparison for a 136# CWR wood tie 5 deg curve track
with FRA Class 4 line defects when the lateral resistances for “weak”, “average” and “strong”
conditions are as shown in the right hand side of the figure. The difference in the respective
buckling regimes is evident. In fact, for the weak resistance condition there is no buckling regime
indicating a highly unstable condition as evidenced by the progressive lateral displacements with
temperature. Thus the inference from the figure is that for this track type/condition the “strong”
resistance track buckles in between 100 – 140°F above neutral, “average” resistance between 87
and 94°F, and “weak” resistance track does not buckle in the conventional sense, but
progressively shifts with temperature, indicating a highly unstable and unsafe condition.
Figure 15 – Lateral Resistance Influence on Bucklin Temperatures for Weak, Average and Strong Track Conditions
wood tie track136RE rail5 deg curve20 in. tie spacingGranite, µ=1.20Kv=6000 psiKf=200 psiτo=120 in-kips/rad/in
δo=1.5 in. Lo=180 in
Input parameters
0.25
2000
3600 Strong LateralResistance
1600
0.35
2400
0.5
1400
Average LateralResistance
Weak LateralResistanceT
em
pe
ratu
re in
crea
se a
bo
ve n
eu
tra
l (°F
)
0 2 4 6 8 10 12 14 160
20
40
60
80
100
120
140
Lateral displacement (in)
Large BucklingRegime
Small BucklingRegime
Strong LateralResistance
Average LateralResistance
Weak LateralResistance
(No Buckling Regime)
wood tie track136RE rail5 deg curve20 in. tie spacingGranite, µ=1.20Kv=6000 psiKf=200 psiτo=120 in-kips/rad/in
δo=1.5 in. Lo=180 in
Input parameters
wood tie track136RE rail5 deg curve20 in. tie spacingGranite, µ=1.20Kv=6000 psiKf=200 psiτo=120 in-kips/rad/in
δo=1.5 in. Lo=180 in
Input parameters
0.25
2000
3600
0.25
2000
0.25
2000
3600 Strong LateralResistance
1600
0.35
2400
1600
0.35
2400
0.5
1400
Average LateralResistance
Weak LateralResistanceT
em
pe
ratu
re in
crea
se a
bo
ve n
eu
tra
l (°F
)
0 2 4 6 8 10 12 14 160
20
40
60
80
100
120
140
Lateral displacement (in)
Large BucklingRegime
Small BucklingRegime
Strong LateralResistance
Average LateralResistance
Weak LateralResistance
(No Buckling Regime)
© 2011 AREMA ®
(b) Influence of Lateral Resistance on “Safe” Temperatures for Buckling Prevention
As discussed, lateral resistance is a highly variable parameter, and for concrete ties it typically
falls into the four peak resistance ranges: (1) less than 1900 lbs/tie for weak, recently maintained
conditions; (2) 1900 to 2500 for marginal conditions where consolidation is active; (3) 2500 to
3200 for average conditions, and (4) above 3200 for well maintained, fully consolidated tracks.
Figure 16 shows the influence of these 4 categories on “safe rail temperatures” for buckling
prevention for tangent and curved tracks with FRA Class 4 line defects and with a neutral
temperature of 60°F. Also shown on the figure is the measured data form the Volpe/AMTRAK
tests from Figure 9.
Figure 16 – Influence of Lateral Resistance on Safe Temperature Increase
Figure 16 indicates the importance and benefit of consolidation in terms of safe rail temperatures
by increasing it from a low 118°F to a more acceptable 134° for the 5 deg curve, and from 138°F
to 144°F for the tangent. Note that for higher neutral temperatures, the safe temperature values
increase correspondingly by the amount of the neutral temperature change.
Parameters: Tangent and 5° (350m radius) concrete tie track with Class 4 (38mm/10m) linedefect; 136# CWR; TN=60°F (16°C); variable lateral resistances
Safe rail temperature, Tall
90 100 110 120 130 140 150 160
1400
1800
2200
2600
3000
3400
3800
La
tera
l re
sis
tan
ce (
lbs
/tie
)
TN = 60°F(16°C)5 deg curve
5
10
15
La
tera
l res
ista
nc
e (k
N/tie
)
°F
°C40 50 60 7030
TypicalConsolidation Regime
Strong
Average
Weak
TN = 60°F(16°C)Tangent
Volpe/AMTRAKNew Carrollton Data
Before Surfacing
After DTS
After Surfacing
Parameters: Tangent and 5° (350m radius) concrete tie track with Class 4 (38mm/10m) linedefect; 136# CWR; TN=60°F (16°C); variable lateral resistances
Safe rail temperature, Tall
90 100 110 120 130 140 150 160
1400
1800
2200
2600
3000
3400
3800
La
tera
l re
sis
tan
ce (
lbs
/tie
)
TN = 60°F(16°C)5 deg curve
5
10
15
La
tera
l res
ista
nc
e (k
N/tie
)
°F
°C40 50 60 7030
TypicalConsolidation Regime
Strong
Average
Weak
TN = 60°F(16°C)Tangent
Parameters: Tangent and 5° (350m radius) concrete tie track with Class 4 (38mm/10m) linedefect; 136# CWR; TN=60°F (16°C); variable lateral resistances
Safe rail temperature, Tall
90 100 110 120 130 140 150 160
1400
1800
2200
2600
3000
3400
3800
La
tera
l re
sis
tan
ce (
lbs
/tie
)
TN = 60°F(16°C)5 deg curve
5
10
15
La
tera
l res
ista
nc
e (k
N/tie
)
°F
°C40 50 60 7030
TypicalConsolidation Regime
TypicalConsolidation Regime
Strong
Average
Weak
TN = 60°F(16°C)Tangent
Volpe/AMTRAKNew Carrollton Data
Before Surfacing
After DTS
After Surfacing
Before Surfacing
After DTS
After Surfacing
© 2011 AREMA ®
5.0 Conclusions and Recommendations 1. Track lateral resistance is a key parameter for lateral stability assurance. It is typically
represented through “spring” type load-deflection characteristics which are non-linear and highly
variable. The key descriptors are the “peak” and “limit” values which depend on: tie type, weight,
shape and spacing; ballast type/condition, shoulder width, crib content; ballast maintenance,
degree of consolidation, and train loads. It is measured most conveniently by a single-tie-push-
test (STPT) method which mobilizes the tie in the ballast and records the load-deflection
response.
2. Lateral resistance has both static and dynamic components. The dynamic characteristics
depend on the loaded and unloaded (dynamic uplift) frictional interface between tie and ballast as
represented by a tie/ballast friction coefficient parameter which is experimentally determined. It is
this dynamic resistance value that’s required in lateral stability analyses.
3. Ballast maintenance has a large influence on reducing lateral resistance which can be on the
order of 30-70%. Such reductions can cause damaging track failures through track buckling or by
excessive misalignments generated by track shift. In order to mitigate such failures, lateral
strength restoration is required through either mechanical (DTS) or train traffic (MGT) induced
consolidation.
4. Whereas DTS has proven to be an effective and efficient means to quickly restore 30-50% of
the reduced resistance, the train traffic (MGT) based consolidation may require tonnage levels
well in excess of the currently accepted 0.1 MGT equivalent, as US data indicates. Since even
the 0.1 MGT’s slow order traffic may be prohibitive for most railroads, more tests and analyses
are required to better evaluate the tonnage consolidation aspects. Such studies need to address
several ballast compaction mechanics issues, including the effects of axle loads and speeds on
© 2011 AREMA ®
consolidation, track curvature influences, track type/material/component influences, and recovery
rates required for lateral stability assurance.
5. Track lateral stability assurance requires addressing the two key failure modes of track
buckling and track shift. The track buckling safety limits in terms “allowable temperature increase
values” above neutral are strongly lateral resistance dependent, as are the actual buckling
temperatures. Hence restoration of reduced resistance due to maintenance to “safe” levels is a
key part of track buckling prevention. Such lateral strength restorations through dynamic track
stabilization (DTS) have proven to be quite effective, whereas the similar effects through tonnage
(MGT) require more evaluation and testing.
6. Mitigation of track shift and resulting unsafe alignment defects also require high lateral
resistances. The particular requirement is a high lateral resistance elastic limit, which can be
achieved through a high peak resistance value. The impact of such high elastic and peak values
is an allowance for higher net axle L/V’s and higher speeds.
REFERENCES
1. Samavedam, G. and A. Kish, “Continuous Welded Rail Track Buckling Safety Assurance
through Field Measurement of Track Resistance and Rail Force”, Transportation Research
Record 1289, pp.39-52
2. Kish, A. and G. Samavedam, “Track Buckling Prevention: Theory, Safety Concepts and
Applications”, US DOT/FRA Report, 2004, in FRA publication
3. Lichtberger, B, “Track Compendium”, Eurailpress, 2005, English Edition
© 2011 AREMA ®
4. Bosch van den, R, “Quervershiebewiderstandsmessung mit dem dynamischen
Gleisstabilitator“ in German, Eisenbahningenieur (58) 6/2007, pp. 15-19
5. Kish, A., G. Samavedam, and D. Wormley, “New Track Shift Safety Limits for High-Speed
Rail Applications”, Proceedings of World Congress on Railway Research, Cologne, Germany,
November 2001
6. Kish, A.,” New Track Shift Limits for High-Speed Rail” in French, German and English,
Schienen der Welt – Rail International, August-September, 2001, pp 36-41
7. Hunt, D and Z. Yu, “Measurement of Lateral Resistance Characteristics for Ballasted
Tracks”, ERRI D 202/DT 361 Report, February 1998, Utrecht, Netherlands
8. Sluz, A., TSC/Volpe Project Memorandum, 1985
9. Trevizo, C., “Restoration of Post Tamp Stability”, WP-150, AAR/TTCI Project Report, 1990
10. Kish, A., D. Clark, and W. Thompson, “Recent Investigations on the Lateral Stability of
Wood and Concrete Tie Tracks”, AREA Bulletin 752, Volume 96, October 1995
11. Sluz, A., “Evaluation of Dynamic Track Stabilization”, Volpe Project Technical
Memorandum, 2000
12. Kish, A., T. Sussmann, M. Trosino, “Effects of Maintenance Operations on Track Buckling
Potential”, Proceedings of International Heavy Haul Association, May 2003
© 2011 AREMA ®
13. Sussmann, T., A. Kish, M. Trosino, “Investigation of the Influence of Track Maintenance on
the Lateral Resistance of Concrete Tie Track”, Transportation Research Board Conference,
January 2003
14. Samavedam, D., “Track Resistance Measurements on the Union Pacific”, Union Pacific
Test Report, 2001
15. Kish, A and W. Mui, “TREDA – Track Residual Deflection Analysis”, Software and User’s
Guide, March 2001
16. Samavedam, G. and A. Kish, “Track Lateral Shift Model Development and Test Validation”,
DOT/FRA/ORD Report, February 2002
17. Kish, A and G. Samavedam, “Dynamic Buckling of Continuous Welded Rail Track:
Theory, Tests, and Safety Concepts,” Transportation Research Record No. 1289, “Rail: Lateral
Track Stability 1991”
18. Samavedam, G., A. Kish, A. Purple, and J. Schoengart, “Parametric Analysis and Safety
Concepts of CWR Track Buckling”, DOT/FRA/ORD-93/26, December 1993.
19. ERRI D202/DT 360, “Lateral Resistance Tests“, October 1997, Utrecht, Netherlands
20. Kish, A., G. Smavedam and K. Kasturi, “CWR-SAFE Version 2000,” Software and
User’s Guide, CD-ROM (Interim Version), Volpe Center, February 2001
© 2011 AREMA ®
LIST OF FIGURES
Figure 1- Track Lateral Resistance Spring Analogy
Figure 2- Lateral Resistance Load Deflection Characteristics
Figure 3- Typical Peak Lateral Resistance Values
Figure 4- Lateral Resistance Components
Figure 5- Figure 5 – Single Tie Push Test (STPT) Measurement
Figure 6- Dynamic Resistance Concept
Figure 7- Dynamic Resistance Friction Coefficient Behavior
Figure 8- Concrete Tie MGT Consolidation Tests at FAST (Ref. 10)
Figure 9- Concrete Tie Track Stabilization Tests on Amtrak/New Carrollton (Ref. 12)
Figure 10- Track Lateral Shift Parameters and L/V Response
Figure 11- Lateral Resistance Elements for Track Shift Analysis
Figure 12- Elastic Limit Influence on Track Shift Response
Figure 13- Track Buckling Mechanism Stability Behavior
Figure 14- Peak and Limit Resistance Impacts on Buckling Temperatures
Figure 15- Lateral Resistance Influence on Bucklin Temperatures for Weak, Average
and Strong Track Conditions
Figure 16- Influence of Lateral Resistance on Safe Temperature Increase
© 2011 AREMA ®
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Andrew Kish, Ph.D.Kandrew Inc. Consulting Services
Peabody, MA, USA
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
q Paper deals with track lateral resistance issues addressing track lateral stability management including maintenance, geometry retention and track buckling prevention.
Ø What is track resistance and key influencing parameters?Ø What impacts on track maintenance, track buckling, and track
shift?
Paper Content
• Track lateral resistance fundamentals• Lateral resistance and ballast stabilization• Lateral resistance influence on track stability• Key research needs for more effective management
Key Issues Addressed
Overview
Paper Overview, Content, and Key IssuesConsulting Services
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Talking Points
• What is track lateral resistance, key components/parameters and how to measure?
• What maintenance aspects/requirements for consolidation?
• What US data on resistance restoration after ballast work?
• What impacts on track buckling and track shift?
• What research needs?
Presentation OutlineConsulting Services
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
1) High L/V’s2) Reduced local lateral resistance3) Initial imperfections (welds), construction
anomalies, and install errors
1 Formation of initialtrack misalignments
1) Increase in L/V, and high longitudinal forces2) Reduced lateral resistance at line defects3) Track “dynamic uplift”4) Many cycles of L/V’s
2Growth of
misalignments(Track Shift)
1) High longitudinal force2) Reduced TN (stress-free temperature)3) Weakened lateral resistance4) Train loads and dynamics5) Misalignments generated by track shift
3 Buckling
Event Major causal factorsStage
Track Lateral Stability Mechanism
What is track lateral stability?
0.0
0.1
0.2
0.3
0.4
0.5
Cu
mu
lati
ve
late
ral
resid
ual d
efl
ecti
on
(in
)
0 100 200 300 400 500Number of axle passes
Low L/VModerate L/V
High L/V
High Axle Load Problem
TREDA
High Thermal Load Problem
CWR-SAFE
Consulting Services
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
TLR is the reaction offered by the ballast to therail-tie structure against lateral movement
g Definition:
Consulting Services
What is track lateral resistance (TLR)?
Ballast reaction (lb/tie)or (lb/in)
PP
P
P
P
P
Lateral spring
L
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
How to Measure?
• single tie push test (STPT)
• discrete cut panel pull test
• continuous track panel pull test (TLPT)
• continuous dynamic measurement (Plasser-DTS)
• analytic empirical model (CWR-SAFE)
Measurement Methods
• single tie push test (STPT)
• discrete cut panel pull test
• continuous track panel pull test (TLPT)
• continuous dynamic measurement (Plasser-DTS)
• analytic empirical model (CWR-SAFE)*
√
* Model “trained” by over 1000 STPT measurements to provide lateralresistance based on inputs of: tie type, shoulder width, crib content, and consolidation
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
4000
Ap
plied
Lo
ad
(lb
s)
Limit Resistance
weak
Initial Stiffness average
Peak Resistance
strong
Consulting Services
Typical Behavior and Values
0 1 2 3 4
Displacement (in)
0
500
1000
1500
2000
2500
3000
3500
Ap
plied
Lo
ad
(lb
s)
Limit Resistance
Initial Stiffness
Peak Resistance
Idealization
Typical concrete tie peak values (static)
1800 2000 2200 2400 2600 2800 3000 3200 3400 3600
Weak Marginal* Average Strong
7.5 10.0 12.5 15.0
lbs/tie
kN/tie
Note: for timber ties subtract 500 lbs/tie * Typical consolidation range
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
§ Tie type, weight, shape and spacing; ballast type andcondition (fouled, frozen, etc.) shoulder width, crib content; maintenance, degree of consolidation, and vehicle loads (dynamic uplift)
Consulting Services
Factors Influencing Lateral Resistance
§ Tie type, weight, shape and spacing; ballast type andcondition (fouled, frozen, etc.) shoulder width, crib content; maintenance, degree of consolidation, and vehicle loads (dynamic uplift)
§ Tie type, weight, shape and spacing; ballast type andcondition (fouled, frozen, etc.) shoulder width, crib content; maintenance, degree of consolidation, and vehicle loads (dynamic uplift)
Dynamic Uplift
Reduced ResistanceIncreased Resistance Increased Resistance
25%
35%
40%
Can result up to 40% loss of lateral resistance
% Contribution Rule of Thumb
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
Track Dynamic Resistance: Tie/Ballast Friction Coefficient
Fdyn = Fstat + µRV
RV
Fdyn
RV
Friction coefficient, µ, is a tie bottom roughness index
• Friction coefficient µ is a key parameter in both track buckling and track shift analyses
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
• Ballast maintenance can reduce TLR by 40 - 70%; requires consolidation either by dynamic track stabilization (DTS) or traffic. DTS can increase value by 30 – 40%; traffic consolidation may require over 0.1 MGTs (million gross tons) of traffic coupled with slow-orders.
Vertical load (1400 psi hydraulic pressure)
Horizontal vibration (30 Hz)
Speed: 1mph
DTS principle: a quick restoration of the tie bottom friction component of lateral resistance by applying vertical load coupled with horizontal vibration
Consulting Services
Influence of Ballast Maintenance
• Initial assumption on traffic consolidation: 0.1 MGT = 30-40% DTS
Has not been demonstrated by US experience!
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
US Data on Ballast Consolidation
Chessie and Amtrak Tests – concrete ties (1985, Sluz, [8])• Chessie: 9% recovery after 0.076MGT• Amtrak: 11% recovery after 0.073MGT
AAR/TTC Tests – wood ties (1990 – Trevizo, [9])• Tangent: 17% recovery after 0.1MGT; 32% after 1MGT• 5° Curve: 9% recovery after 0.1MGT; 21% after 1MGT
Volpe/FRA Tests at TTC - wood and concrete (1987-1990; Kish, et al, [10])• Wood - tangent: 26% recovery after 0.1MGT• Concrete – 5° Curve: 52 % reduction due to tamping• Concrete – 5° Curve: 22% recovery after 0.1MGT
Volpe/Union Pacific Tests – concrete (2000-Sluz, [11])• Concrete (new-scalloped) - 17% recovery after 0.35MGT • DTS increase - 33%
Volpe/Amtrak/FRA Tests – concrete (2001- Kish, et al, [12, 13])• 43% reduction due to surfacing with ½ inch lift• DTS increase: 31%
UP/Foster-Miller Tests – wood/concrete/old/new/DTS (2001-Samavedam [14])• 39-70% reduction due to tamping/surfacing• 0.1MGT had negligible influence on wood after tamping; 0.2MGT was 28%• DTS increase on concrete: 22%• After DTS, traffic decreased lateral resistance initially, then it increased• Heavy rain/wet ballast: 20% decrease in lateral resistance on wood
Chessie and Amtrak Tests – concrete ties (1985, Sluz, [8])• Chessie: 9% recovery after 0.076MGT• Amtrak: 11% recovery after 0.073MGT
AAR/TTC Tests – wood ties (1990 – Trevizo, [9])• Tangent: 17% recovery after 0.1MGT; 32% after 1MGT• 5° Curve: 9% recovery after 0.1MGT; 21% after 1MGT
Volpe/FRA Tests at TTC - wood and concrete (1987-1990; Kish, et al, [10])• Wood - tangent: 26% recovery after 0.1MGT• Concrete – 5° Curve: 52 % reduction due to tamping• Concrete – 5° Curve: 22% recovery after 0.1MGT
Volpe/Union Pacific Tests – concrete (2000-Sluz, [11])• Concrete (new-scalloped) - 17% recovery after 0.35MGT• DTS increase - 33%
Volpe/Amtrak/FRA Tests – concrete (2001- Kish, et al, [12, 13])• 43% reduction due to surfacing with ½ inch lift• DTS increase: 31%
UP/Foster-Miller Tests – wood/concrete/old/new/DTS (2001-Samavedam [14])• 39-70% reduction due to tamping/surfacing• 0.1MGT had negligible influence on wood after tamping; 0.2MGT was 28%• DTS increase on concrete: 22%• After DTS, traffic decreased lateral resistance initially, then it increased• Heavy rain/wet ballast: 20% decrease in lateral resistance on wood
0.1 MGT traffic does not
produce the DTS equivalent,
but LOWER!
More R&Dand tests are
required to better quantify tonnage
based consolidation
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
What resistance is required?
(1) The % recovery is important, but MORE important are the actual values i.e. wherefrom and where to?
(a 35% recovery from 1800 lbs/tie is different than a 35% recovery from 2300 lbs/tie)
(2) What is the minimum resistance required for buckling safety?
Answer: depends on track parameters and conditions!
• Track curvature
• Track alignment
• Track neutral temperature
Need to know/measure absolute values!
Key Notes/Questions
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Parameters: 5° (350m radius) curve; concrete tie track with Class 4 (38mm/10m)line defect; US-136# CWR; variable lateral resistances CWR-SAFE
Consulting Services
Minimum Resistance for Buckling Safety
90 100 110 120 130 140 150 160
1400
1800
2200
2600
3000
3400
3800Strong
Average
Marginal
Safe allowable rail temperature
Late
ral re
sis
tan
ce (
lbs/t
ie)
Weak
TN = 60°F(16°C)
TN = 70°F
TN = 90°F
TN = 80°F
TN = 100°F (38°C)
TN = 50°F
5
10
15
Late
ral re
sis
tan
ce (k
N/tie
)
°F
°C40 50 60 7030
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Parameters: Tangent and 5° (350m radius) concrete tie track with Class 4 (38mm/10m) line defect; 136# CWR; TN=60°F (16°C); variable lateral resistances
Safe rail temperature, Tall
90 100 110 120 130 140 150 160
1400
1800
2200
2600
3000
3400
3800
La
tera
l re
sis
tan
ce (
lbs/t
ie)
TN = 60°F(16°C)5 deg curve
5
10
15
La
tera
l resis
tan
ce (k
N/tie
)
°F
°C40 50 60 7030
TypicalConsolidation Regime
Strong
Average
Weak
TN = 60°F(16°C)Tangent
Consulting Services
Influence of Track Stabilization on Buckling Safety
Lateral deflection, δ
∆Τ
TBmin
T Bmax
Buckling Regime
BUCKLE
Lateral deflection, δ
∆ T
T Bmin
T Bmax
BucklingRegime re
BUCKLE
Track quality: MARGINAL
Track quality: AVERAGE
(Safe Temp Increase)
Realized Benefit Is MORE!!
Before Surfacing
Volpe/AMTRAKNew Carrollton Data
After DTS
After Surfacing
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
§ Track Shift: incurrence of cumulative residual deflections undermany axle L/V (NAL) passes
Ø Moving axle loads; vertically loaded and unloaded lateral resistance;thermal loads; curvature and alignment defects
Consulting Services
Influence of Lateral Resistance Track Shift
xo
V
L
V
L
L
Deflection
Res. Defl.
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Tack Shift Residual Deflection Mechanism: Moving L/V Loads
xo
δresidual
Consulting Services
Track Shift and Moving Loads
• Net residual deflections are a cumulative sum of axle load passes!
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
Lateral Resistance Influence on Track Stability
0 1 2 3Displacement, w (in)
Lateral Displacement
TR
TN
Bucklingregime
Tall
Tbmax
Tbmin
0
500
1000
1500
2000
2500
3000
3500
La
tera
l R
esis
tan
ce
, F
(lb
s) Peak Resistance, FP
Limit Resistance, FL
WP WL
Track Buckling
Fp (stat)
Unloaded tie
Fe
WeWp
Lateral Deflection, W
Fp (dyn)
µ • Rv
Loaded tie
Tie ballastfrictioncoefficient
Tie load
Track Shift
0 10 20 30 40 50 60 70 80 900.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Resi
dual d
eflect
ion
(in)
Number of axle passes
We=0.005 in.
We=0.03 in.
We=0.05 in.
We=0.015 in.
2° concrete tie; 136#;
ÎT=50°F;V=55 kips;
FP=2000lbs; Fe=1500lbs; wp=0.3in; L/V=0.5
0 10 20 30 40 50 60 70 80 900.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Resi
dual d
eflect
ion (
in)
Number of axle passes
We=0.005 in.
We=0.03 in.
We=0.05 in.
We=0.015 in.
2° concrete tie; 136#; ÎT=50°F;V=55 kips; FP=2000lbs; Fe=1500lbs; wp=0.3in; L/V=0.5
Elastic Limit Influence
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
Lateral Resistance Influence on Track Stability
Key Track Shift Issues
FRA Vehicle/Track Interaction Safety Limits on Track Shift
L/V = 0.4 + 5/V(L/V is net axle ratio)
L66=31.4 kips or (L/V)66= 0.48
(what lateral resistance is required tomeet criterion?)
What is the allowable net axle L/V for high speed passenger operation?
What are the limiting L/V conditions for heavy freight/long train applications?
Current R&D Need
Requires industry driven R&Dprogram to evaluate limiting
conditions/factors
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
Conclusions
• Track lateral resistance is an important parameter for track geometry retention and trackstability management
• It is a complex parameter: highly non-linear, variable, difficult to measure, has both static and dynamic components where the tie/ballast friction coefficient plays a key role
• Ballast work (surfacing, lifting, tamping) reduces lateral resistance considerably (40 -70%) requiring quick and efficient restoration HOW TO RESTORE?
• Dynamic track stabilization (DTS) has proven to be a quick and effective means to restore lateral resistance, and typically does not require any speed restrictions.
• Based on US data, traffic (MGT) consolidation for equivalent lateral resistance recovery may require larger than the currently accepted 0.1MGTs. MORE R&D IS REQUIRED FOR EVALUATION!
• Track lateral resistance is a key parameter for both track buckling and track shift evaluations, but each requires different components of the lateral resistance function
Where to?
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
Consulting Services
Challenges/Research Needs
Key Research Need: quantification of MGT consolidation behavior i.e. tonnage requirements for ballast resistance recovery:
qWhat is the mechanics of ballast compaction under traffic loads?
ØWhat is the influence of axle loads?
ØWhat is the influence of train speeds?
ØWhat is the influence of track types/conditions?(concrete/wood/ tangent/curved)
ØWhat is the influence of reduced resistance on recovery?(initial conditions)
Bottom Line: is the 0.1 MGT adequate? If NOT, what is??
2011 ANNUAL CONFERENCESeptember 18-21, 2011 | Minneapolis, MN
There’s got tobe a better way!