Escolar Documentos
Profissional Documentos
Cultura Documentos
Abstract: Current practices in railway track analysis and design are reviewed and discussed in this paper. The
mechanical behavior of railway track structure comprising of various components has not been fully understood due
to the railway track structural complexity. Although there have been some improvements in the accuracy of current
track design methods in recent decades, there are still considerable uncertainties concerning the accuracy and
reliability of the current methods. This indicates a need for a thorough review and discussion on the current practices
in the analysis and design of railway track systems. In this paper, railway design approaches proposed by various
standards along with the results of a wide range of technical researches are studied and necessary suggestions are
made for the improvement of current practices in the analysis and design of railway tracks.
Keywords: railway track, track components, analytical models, design criteria, track design procedure.
BR M
reliability of the Winkler model. As a result, more Ps g
realistic approaches are proposed and some V
India M 1
improvements are made in the basic beam on 58.14 u 0.5
elastic foundation model. For instance, beam V
South Africa M 1 4.92
on discrete support model has been developed D
19.65V
and more recently analyzed using a practical Clarke M 1 1
Du 2
energy approach [3] to compensate for the errors
WMMTA M (1 3086 * 10 5 V 2 ) 0.67
caused by the assumption of continuous support
under the rail. Pasternak foundation and Sadeghi M 1.098 8 u10 4 V 10 6 V 2
double beam models are also introduced [2] to
take into account the interaction between track stress at the lower edge of the rail head also may
supporting layers and multi-layer nature of track be critical if the vehicles impose high guiding
support, respectively. Kerr [1] has reported the forces between wheel flange and rail head while
result of a research in which the effect of the passing around the curves. Having calculated the
rotational stiffness of the sleepers in calculation magnitude of rail bending stress, comparison
of rail deflection and bending moment are should be made between this stress and the
considered. allowable limit. AREMA [5] has recommended a
Design procedure for a specific rail section practical methodology for calculation and
always starts with the calculation of the design controlling of rail bending stress based upon
wheel load. This load is defined as the product of fatigue consideration and through the
static wheel load and a corrective factor known as determination of several safety factors.
dynamic impact factor to compensate for According to this method, the allowable bending
dynamic as well as impact effects of wheel load stress is defined as:
resulted from wheel and rail surface y t
irregularities. Taking into account various all
(1 A)(1 B )(1 C )(1 D )
(4)
parameters which affect the magnitude of
dynamic impact factor, several researchers along where, (y is the yield stress of rail steel and (t
with railway authorities has been proposed is the longitudinal stress due to temperature
different relationships for the estimation of this changes and can be calculated using the
parameter. Some of these equations are following equation:
summarized in Table 1.
The magnitude of vertical rail deflection Vt E .D .'t (5)
calculated using Equation (1) is greatly
dependent upon track modulus. Track modulus is The parameters A, B, C, and D in Equation (4)
defined as the load required producing unit are safety factors to account for rail lateral
vertical deflection in unit length of the rail. For bending, track condition, rail wear and corrosion,
typical tracks with light to medium rails, and unbalanced superelevation of track,
AREMA [5] recommends a value of 13.8 MPa respectively. Some of the recommended values of
when calculating vertical rail deflection. the above safety factors are presented in Table 2.
The rail bending stress is usually calculated at Wheel-rail contact stresses mainly include
the center of the rail base assuming the pure rolling and shear stresses. The magnitude of these
bending conditions to be applicable. The bending stresses is greatly dependent upon the geometry
1
W all V o W all 0.3V ult (11)
3 mean
Fig. 4. Track deflection criteria for durability [9] Wa Wt .(1 K w .Dc ) .D A (12)
Downloaded from ijce.iust.ac.ir at 7:35 IRST on Friday November 17th 2017
vertical deflection and rail wear life. These Having estimated Wa and considering the
criteria are presented hereunder. maximum rail head limit ( A ) , the rail wear life
AREMA [5] has proposed a limiting range for could be calculated from the following formula:
the magnitudes of vertical rail deflections. A
According to this recommendation, extreme Ty (13)
Wa
vertical rail deflections should be kept within the
range of 3.175 to 6.35 millimeters. Lundgren and Danzig and his colleagues [11] from the
his colleagues [9] has incorporated this AREMA association also carried out extensive
recommendation and proposed the diagram investigations to find a proper formulation for the
presented in Figure 4 as the limit values of estimation of rail wear life. Based on the results
vertical rail deflection. This diagram is based obtained, they suggested the following equation
upon the capability of the track to carry out its which represents the rail wear life in terms of MGT
design task. passed the track over a specific period of time:
Figure 5. Using the diagrams outlined in this the sleeper and ballast [4]. Based on these
figure, maximum allowable rail head height and considerations, various relations are proposed
width loss could be determined. and summarized in Table 3.
The values indicate acceptable rail wear life However, for the purpose of simplification, it
usually range from 20 to 50 percent in rail head would be more practical to consider only the
area reduction. The weight of unit length of the effect of some of the above mentioned
rail, the amount of MGT passed over the track parameters and define, for example, the value of
during its service life, and the train speed are the vertical rail seat load as a function of sleepers
most important parameters which determine the type and spacing. This approach is widely
proper values of allowable rail wear limits to be accepted by many railway authorities. For an
chosen. The more weight of unit length of the
rail, the greater amount of rail head area
reduction would be allowed. Oppositely, the Table 3. Relations for the calculation of rail seat load
greater amount of MGT and higher values of
train speed call for more limited rail head area
reduction.
2.2. Sleeper
The exact contact pressure distribution is possible, it is usually presumed that the
between the sleeper and the ballast and its uniform pressure under the sleeper distribute in
variation with time, will be of importance in the certain portion of the sleeper length (area). This
structural design of sleepers. When track is length (area) is referred to as Effective Length
freshly tamped the contact area between the (Area) and commonly shown with L (Ae) in
sleeper and the ballast occurs below each rail the literature. This assumption is made to
seat. After the tracks have been in service the facilitate the procedure of design calculations.
contact pressure distribution between the sleeper The static equilibrium in vertical direction is then
and the ballast tends towards a uniform pressure applied to acquire the magnitude of contact
distribution [4]. This condition is associated with pressure under the sleeper. A factor of safety is
a gap between the sleeper and the ballast surface also included to account for variations in the
below the rail seat. The most accepted contact sleeper support. Therefore, the average contact
pressure distribution patterns between sleeper pressure between the sleeper and the ballast Pa
and ballast are presented in Table 4. (kPa) can be acquired by:
Although some researches have been
qr
conducted to determine the in-track distribution Pa .F (15)
B.L 3
L L
Developer Description
Ae: Two third of
Table 5. In-track sleeper loading pattern [16] AREMA [12] sleeper area at its
bottom surface
Ae = 6000 cm2 for l =
2.5 m
UIC [13]
Ae = 7000 cm2 for l =
2.6 m
(1)
L (l g)
Australia [14, 15] (2)
L 0.9u( l g )
lg
Schramm [11] L
2
(l g ) (3)
Clarke [11] L ( l g ) 1
125t 0.75
Clarke (simplified) L
l
[11] 3
(1) For bending moment calculation at rail seat
(2) For bending moment calculation at sleeper center
(3) the parameter t is the sleeper height
Schramm
Ballast and sub-ballast layers are composed of
UIC granular materials and used in track structure
Raymond
mainly to sustain the loads transferred from the
sleeper. Other important functions of these layers
Fig. 7. Recommended methods for sleeper bending include: 1) to reduce the stress intensity to the
moment calculation level to be tolerable for subgrade layer, 2) to
absorb impact, noise and vibration induced from
2.3. Rail Fastener the wheels, 3) to restrict the track excessive
settlement, 4) to facilitate track maintenance
Rail fasteners also known as fastening systems operations, especially those related to the
are used in railway track structure to fasten the correction of track geometry defects, and 5) to
rails on the sleepers and to protect the rail from provide adequate drainage for the track structure,
inadmissible vertical, lateral, and longitudinal thereby track settlement as well as vegetation
growth will be limited. In addition to the above numerical solution of the following equation:
mentioned functions, the sub-ballast layer is used
to act as a filter layer which prevents ballast and 3 Pa H A K B d H . dK
Vz
subgrade materials to be mixed together. 2S H A (16)
^( x H ) 2 ( y K ) 2 z2` 2
5
K B
Current practices in structural analysis and
design of ballast and sub-ballast are dealt with the
determination of minimum required depth for It should be noted that, ballast and sub-ballast
theses granular layers. Theoretical, semi- layers are assumed as a single homogenous and
empirical, and empirical methods have been isotropic layer in Boussinesq elastic theory.
developed and are being used in order to satisfy Although such an assumption seems not to be
this design criterion. sufficiently accurate, however, based on the
Theoretical determination of minimum comparison to the available field tests results, the
required depth based on the results of Boussinesq ORE [26] investigations have proved the validity
elastic theory applied to a uniform rectangular of Boussinesq elastic theory. This comparison is
loaded area (see Figure 9) is performed using the presented in Figure 10. It is apparent from this
Fig 10. Comparison of experimental vertical stress Current practices of anlysis and design of
distribution with depth and the Boussinesq solution [26]
railway track systems are entirely based on load
bearing approach in which every track
Considering the 2:1 load spread distribution as component is evaluated separately for its
indicated in Figure 12, the minimum required sufficient strength to sustain the environmental
ballast and sub-ballast depth can be calculated and traffic loads. Various analysis methods and
from the following formula: design criteria have been reviewed in this paper.
Based on comparisons of the available railway
B. L
V z Pa (17) design methods, some shortcomings are
( B z )( L z )
discussed in this section.
Having determined the allowable subgrade Results of the main proposed dynamic impact
load carrying capacity and substituting it in the factors presented in Table 1 are drawn against
equations (16) or (17) the minimum required train speed in Figure 13. It is apparent from this
ballast and sub-ballast depth can be calculated. figure that train speed can considerably influence
[8] Smith, J.O., and Liu, C.K., 1953, Stresses Due [18] American Railway Engineering and
to Tangential and Normal Loads on an Elastic Maintenance of Way Association, Manual for
Downloaded from ijce.iust.ac.ir at 7:35 IRST on Friday November 17th 2017
Solid with Application to some Contact Stress Railway Engineering, Volume 1, Chapter 30,
Problems, Transactions of ASME, Journal of Part 4, Ties Concrete Ties. 2006.
Applied Mechanics, pp. 157-166.
[19] American Railway Engineering and
[9] Lundgren, J.R., Martin, G.C., and Hay, W.W., Maintenance of Way Association, Manual for
A Simulation Model of Ballast Support and the Railway Engineering, Volume 1, Chapter 5, Part
Modulus of Track Elasticity, (Masters Thesis), 9, Track Design Qualification Specifications
Civil Engineering Studies, Transportation for Elastic Fasteners on Timber Cross Ties,
Series, No. 4, University of Illinios, 1970. 2006.
[10] Hay, W.W., Schuch, P.M., Frank, M.W., and [20] Australian Standard, AS 1085.19, Railway track
Milskelsen, M.J., Evaluation of Rail Sections, materials, Part 19: Resilient fastening
2nd Progress Report, Civil Eng., Transportation assemblies, 2002.
Series No. 9, Univ. Illinois Project No. 44-22-
20-332, Univ. Illinois, Urbana, 1973. [21] European Committee for Standardization, EN
13146-1, Railway Applications Track, Test
[11] Danzing J. C., Hay W., and Reinschmidt A., Methods for Fastening Systems, Part 1:
Procedures for analyzing the economic costs Determination of Longitudinal Rail Restraint,
of railway roadway for pricing purposes, 2002.
Volume 1 and 2, Report No. RPD-11-CM-R,
Tops on-line Service Inc, 1976. [22] European Committee for Standardization, EN
13146-2, Railway Applications Track, Test
[12] American Railway Engineering and Methods for Fastening Systems, Part 2:
Maintenance of Way Association, Manual for Determination of Torsional Rail Resistance,
Railway Engineering, Volume 1, Chapter 30, 2002.
Part 1, Ties General Considerations, 2006.
[23] European Committee for Standardization, EN
[13] International Union of Railways, Design of 13146-3, Railway Applications Track, Test
Monoblock Concrete Sleepers, UIC CODE, Methods for Fastening Systems, Part 3:
713 R, 1st Edition, 2004. Determination of Attenuation of Impact
Loads, 2002.
[14] Australian Standard, AS 1085.14, Railway track
materials, Part 14: Prestressed Concrete [24] European Committee for Standardization, EN
Sleepers, 2002. 13146-4, Railway Applications Track, Test
Methods for Fastening Systems, Part 4: Effect
[15] Australian Standard, AS 1085.17, Railway of Repeated Loading, 2002.
Track Materials, Part 17: Steel Sleepers,
2002. [25] European Committee for Standardization, EN
13146-7, Railway Applications Track, Test
[16] Sadeghi, J., 2008, Experimental Evaluation of Methods for Fastening Systems, Part 7:
Accuracy of Current Practices in Analysis and Determination of Clamping Force, 2002.
Design of Railway Track Sleepers, Canadian
Journal of Civil Engineering. Vol. 35, pp. 881- [26] Office of Research and Experiments (ORE),
893. Stresses in the Formation, Question D17,
Stresses in the Rails, the Ballast and the