LONG

WELDED
RAILS
INDIAN RAILWAYS INSTITUTE OF CIVIL ENGINEERING
PUNE
(i)
PREFACE
Long Welded Rail (LWR) has now become synonymous with
modern track structure with a major portion of Indian Railways track
having long welded rails. It is imperative that permanent way men
understand all its facets , be it welding, laying or maintenance so that
full benefits are reaped . With this objective, IRICEN publication on
LWR was printed in 1988 which of course requires revision. This
publication is an updated version with a completely new look incorporating
the latest correction slips and provisions of the LWR Manual.
The publication highlights the evolution of the LWR over the
years with brief references to the research work carried out in RDSO
and foreign railways on various aspects of the LWR. A brief description
of the various SEJ layouts now available, latest provision of LWR on
bridges with comments on the state of art, neutral temperature and its
measurement are also included. It is hoped that this publication will go
a long way in helping track engineer to understand the intricacies involved
in laying and maintaining LWR track.
This book has been authored by Shri Ajit Pandit, Sr. Professor
& Dean of this Institute. If there are any suggestions or discrepancies,
kindly write to the undersigned.
Shiv Kumar
Director
IRICEN
(ii)
ACKNOWLEDGEMENTS
While covering the subject on Long Welded Rails at IRICEN
the absence of an updated publication on the subject covering the
state of art and latest instructions was acutely felt. The publication
printed in 1988 required revision to incorporate the provisions of the
LWR Manual 1996, including the latest correction slips.
This IRICEN publication is a result of the desire to fill the
gap and produce a documentation which would be useful for all
practicing civil engineers on Indian Railways.
It would be appropriate to mention the support and
assistance rendered by IRICEN faculty and staff in preparing this
publication. Special mention may be made of Shri Sunil Pophale,
Head Draftsman who rendered valuable assistance in preparing the
drawings. Shri Dhumal, PA assisted in editing the manuscript.
Shri R.K. Verma, Senior Professor/Track gave valuable suggestions
from time to time.
Above all, the author is grateful to Shri Shiv Kumar,
Director/IRICEN for his encouragement and guidance for preparing
the document.
Ajit Pandit
Senior Professor & Dean
(iii)
INDEX/CONTENTS
CHAPTER I : PAGE NO.
INTRODUCTION TO LONG WELDED RAILS 1-8
1.1. Evolution of Long Welded Rail
1.2 Some Basic Definitions
1.3 An Explanatory Note on the
Short Welded Rail
1.4 Advantages of Long Welded Rail
CHAPTER II :
PRINCIPLES OF LONG WELDED RAIL 9-27
2.1 Basic Principles
2.2 Force Diagram
2.3 Importance of rail temperature
2.4 Breathing Length
2.5 Longitudinal Ballast Resistance
2.6 Lateral Ballast Resistance
CHAPTER III :
THERMAL MOVEMENTS AND HYSTERESIS 28-46
3.1 Estimation of Thermal Movements
3.2 Switch Expansion Joints
3.3 Phenomenon of Hysteresis
3.4 Gap measurements at
Switch Expansion Joints
CHAPTER IV :
PERMITTED LOCATIONS AND 47-77
TRACK STRUCTURE
4.1 General Considerations
4.2 Alignment
4.3 Gradients
4.4 Track structure
4.5 LWR on bridges
(iv)
(v)
CHAPTER V : PAGE
NO.
LAYING AND MAINTENANCE 78-95
5.1 Laying of an LWR
5.2 Destressing
5.3 Regular Maintenance operations
5.4 Special Track Maintenance
5.5. Deep screening in LWR territory
CHAPTER VI :
UNUSUAL OCCURENCES IN LWR, 9 6 -
104
INSPECTION & RECORD KEEPING
6.1 Introduction
6.2 Fractures
6.3 Buckling
6.4 Inspection and Record keeping
6.5 Duties, responsibilities and training
CHAPTER VII :
105-128
SPECIAL TOPICS
7.1 Buckling Phenomena
7.2 Tests by German Railways
7.3 Studies conducted by British
Transport Commission
7.4 Static and Dynamic Buckling
7.5 Dynamic Track Buckling Model
7.6 CWR Safety Assurance Program
7.7 Field determination of lateral
ballast resistance
7.8 Neutral Temperature, its variation
and determination
List of References 129
CHAPTER - I
INTRODUCTION TO LONG WELDED RAILS
1.1 Evolution of Long Welded Rails
1.1.1 The dream of a jointless track has fascinated track
engineers ever since the first railway was laid and
efforts are continuing even today for achieving a
fully continuous track without joints.To achieve this
various innovative techniques have been used over
the years. Two ways to achieve this are:
(1) Increasing the length of rails by rolling longer
rails.
(2) Welding the rail joints.
1.1.2 At the beginning of the last century the standard
length of rail was generally 12m. Apart from the
logistic considerations of rail transportation and its
loading and unloading, the length of the rail was
governed by the length of cooling boxes in the rail
manufacturing steel plant, as controlled cooling after
the rolling process was necessary. Subsequent
advancements in the manufacturing process have
enabled rolling of longer rails possible. 26m long
rails are now being manufactured by SAIL in the
Bhilai Steel Plant. Rails of 65m and 78m and even
longer rails upto 480m are being planned for
manufacture in the near future in the steel plants
across the country. Jindal Steel & Power Ltd are
setting up a steel plant in Raigad dist. of
Chhattisgarh state, where longer length rails are
likely to be manufactured.
1.1.3 Although welding of rails was started in 1905 itself,
commercial welding on any considerable scale
became common only after 1932. During the thirties,
(1)
the weights and lengths of standard rail sections
varied from 22kg/m to 65 kg/m and from 5.5 m to
27m respectively (Reference1). The length of welded
panels varied from 18m to 380m. The welding
process most commonly used at that time was
‘thermit' though flash butt and other processes were
in use in America, South Africa and elsewhere.
1.1.4 Although longer length rails were being used, track
engineers were unsure of the expansions which
these longer length rails would undergo under
temperature variations and consequently, the
expansion gaps required at the ends of these
panels. Studies by the British Transport Commission
(Reference 2) however,indicated that the gap to be
provided is not proportional to the length of the
welded rail and in fact the expansion gaps are
actually independent of the length of the rail. What
this implies is that in long welded panels, free rail
expansion or contraction is not being permitted and
thereby undergo what are called as 'thermal
stresses' due to constrained expansion and
contraction. The ability of the welded panel to
withstand compressive stress buildup due to
constrained expansion was another source of worry
to the early track engineers. The British Transport
Commission conducted a number of studies in the
1950s to go into the aspect of lateral stability of the
Long Welded Rails under temperature induced
compressive stresses. Interestingly it was found
that these welded panels derive resistance to
buckling not only from the stiffness of the rail itself
but also from the rail sleeper fastenings and ballast
resistance. It was only after the lateral stability of
rails under compressive forces was confirmed was
the concept of Long Welded Rails accepted by track
engineers.
(2)
1.1.5 In India in the nineteen thirties, the GIP Railway had
undertaken welding of rail joints using the electrical
process. In 1939, BN Railway started conducting
trials on welded rail joints. Between 1940 and 1946,
other railways viz. NW, GIP and EI railways also
commenced trials with welded joints. From 1947 to
1966 large number of 5-rail panels (65m in BG and
60m in MG) and 10-rail panels (130m in BG and
120M in MG) were put into track. The purpose was
to reduce the maintenance efforts by reducing the
number of joints. However, large scale maintenance
problems were reported by various railways
regarding the behaviour of 5-rail and 10-rail panels
due to:
i) increased rail battering and hogging;
ii) elongated fish-bolt holes;
iii) bent fish-bolts.
The phenomena of battering and hogging of rail joints is
shown in Fig. 1.1
Fig. 1.1
Batter
Hog
Rail Rail
Scale
Rail
(3)
→ ←
(4)
SWR
SWR
Fig. 1.2(a)
SWR
LENGTH ≤ 39m in BG
≤ 36m in MG
SPECIAL ARRANGEMENT
FOR TRANSFERING FORCE FROM
STOCK RAIL TO TONGUE RAIL
AND WING RAIL TO CROSSING

CWR
SEJ
BREATHING
LENGTH
CENTRAL PORTION BREATHING
LENGTH
TOTAL LENGTH OFLWR
SEJ
LWR
Fig. 1.2(b)
CENTRAL PORTION
SEJ
SEJ
fCWR CWR g
CWR
Fig. 1.2(c)
Minimum length of LWR
= 250m in BG
= 500m in MG
MAX. LENGTH : ONE BLOCK SECTION LENGTH
Taking cognizance of these problems, the Railway Board
vide their letter No 65/W6/WRL/6 of 20th January
1966 appointed a committee consisting of 3
Engineers in Chief (Track), Joint Director Standards
(Track), RDSO and a Deputy Chief Engineer of
W.Rly to investigate into the behaviour of 5-rail
panels and 10-rail panels at the first instance and
thereafter of Long Welded Rails.
1.1.6 The findings and recommendations of the committee
make interesting reading (Reference1). The
committee found that the IRS fishplate design as
per current standards is inadequate to cater to the
expansion and contraction occuring in 5-rail and 10-
rail panels. While the capacity of the IRS fishplate
design is to accommodate a movement of 15 mm,
the actual movements of a 5 rail or 10 rail panel are
much larger resulting in large gaps, bent fishbolts
and elongated fishbolt holes. The 3-rail panel
therefore appears to be roughly the longest rail
which could be laid in the track with the
conventional fish plated joints. The committee,
therefore, recommended that:
1) Welding of 5-rail and 10-rail panels was to
be discontinued,
2) Existing 5-rail and 10-rail panels were to be
cut into two and half rail panels;
3) RDSO was to conduct further studies for
deciding the track structure, temperature
and ballast conditions for laying the LWR.
1.2 Some Basic Definitions
1) SWR : A rail which expands and contracts
throughout its length: 3-rail panels, 39m in
BG and 36m in MG, will behave like an
SWR.(Fig1.2(a))
2) LWR : (LWR) is defined as a welded rail
whose central portion does not exhibit any
(5)
longitudinal movement on account of
temperature variations. The movements are
exhibited by track lengths on either side of
the central portion which are called as
‘breathing lengths’. While the maximum
length of an LWR is restricted to one block
section length, the minimum length of an
LWR should be 500m in MG and 250m in
BG. (Fig 1.2(b))
3) CWR : An LWR which continues through
station yards including points and crossings
is described as a continuous welded rail.
(Fig 1.2(c))
4) Rail Temperature : The temperature of the
rail as measured by an approved rail
temperature measurement device.
5) Breathing length : The lengths at the ends
of an LWR on either side where the LWR
"breathes" i.e. exhibits movements due to
temperature variations is called the
breathing length.
6) Destressing Temperature (t
d
): The average
rail temperature at the time of fixing the
rail fastenings after a destressing operation
performed manually is called the
destressing temperature.
7) Stress-free Temperature or Neutral
Temperature (t
n
) : The rail temperature at
which an LWR is free of longitudinal thermal
stress is called the stress-free or neutral
temperature.
8) Switch Expansion Joint (SEJ) : A physical
device placed at the end of the breathing
length of an LWR to accomodate the
expansion and contraction of the breathing
length.
9) Buffer rails : A set of free rails used in lieu
of SEJs as a temporary measure to
accomodate the movement of the breathing
length of the LWR.
(6)
1.3 An Explanatory Note on the Short Welded Rail
Indian Railways have adopted a 3-rail panel (39m in BG and
36 m in MG) as the Short Welded Rail(SWR) with fish plated joints
due to reasons mentioned in 1.1.6. Although the SWR is exhibiting
expansion and contraction throughout its length, it is not a free
expansion or contraction due to the action of ballast and fastenings.
The SWR therefore develops thermal forces because of being
subjected to constrained expansions and contractions. It has been
estimated that the magnitude of thermal force in an SWR may be as
high as 70% to 75% of the thermal force in an LWR. This coupled
with fish plated joints makes the SWR very vulnerable and should
therefore be maintained with care. The provisions of laying and
maintenance of SWRs is given in para 505 of the Indian Railways
P-way Manual (1986).
1.4 Advantages of the Long Welded Rail
Today the LWR is synonymous with modern track. The
reason for the widespread popularity of the LWR are the numerous
advantages which are derived using a LWR track structure. The
LWR makes train travel more safe, economical and comfortable due
to the reasons listed below :
l LWR tracks eliminate fishplated joints leading to safety.
Sabotage at fish plated joints has been a major worry
for the Indian Railways and this has again come into
prominence with the accident of the Rajdhani Express at
Rafiqganj in Bihar in which a large number of people
lost their lives and which was attributed to removal of
fishplates on the approach of a bridge.
lFish plated joints are a source of large dynamic forces.
As a result fish plated joints exhibit large scale rail wear
and development of cracks from fishbolt holes and
fractures. In some instances premature rail renewal may
have to be carried out due to excessive fractures.
lDue to development of large dynamic forces at the rail
joints the track geometry at the rail joint gets disturbed
(7)
frequently resulting in an increment in the track
maintenance effort. It has been estimated that there is
as much as 25% to 33% savings in the track repair
and maintenance costs due to elimination of rail joints.
l Due to impact at rail joints there is an added wear and
tear of rolling stock wheels to an extent of 5% and as
the wheel has to negotiate the gap there is added fuel
consumption to an extent of 7%.
l Due to elimination of noise and vibrations at the rail
joints passenger comfort is substantially increased.
(8)
CHAPTER II
PRINCIPLES OF LONG WELDED RAIL
2.1 Basic Principles
A metal rod supported on frictionless rollers can theoretically
expand and contract freely with variations in temperature. If the
length of the metal rod is ‘L’, it will undergo a change in length
equal to L α t, where α is the coefficient of linear expansion and t
is the change in temperature.
For example, if there is a steel rod 13 m long (the same as the
standard length of a rail), then it will undergo change in length
equal to 3 mm, if the temperature of the rod changes by 20
0
C as
can be seen from the calculation below:
Expansion/Contraction=Lα t = (13x1000mm)x(1.152x10
-5
)x20
= 3 mm.
(α =1.152 x 10
-5
per degree centigrade)
However, the rail in the track cannot be compared to the metal
rod supported on frictionless rollers as mentioned in the preceding
paragraph. The rail is restrained from free expansion and
contraction by the sleepers because of,
1. Creep resistance on account of friction between
the rail and sleeper at the rail seat.
2. Creep resistance further offered by the rail-
sleeper fastening.
Thus the expansion or contraction of the rail is less than what it
would undergo if it was completely unrestrained.
In LWR the rail is held down to the sleeper by fastenings which
have adequate toe load, thereby preventing any relative movement
between rail and sleeper. Thus with any change in temperature it
is not the rail alone but the rail-sleeper frame as a whole which
tends to move.
(9)
Here again the rail sleeper frame is not entirely left
unrestrained. The frame is under restraint because of the
resistance offered by the ballast in which the sleepers are
embedded. The resistance offered by the ballast to the movement
of the track frame in the direction of the track is called longitudinal
ballast resistance. Longitudinal ballast resistance is assumed to be
uniform for sleepers of the same type. This longitudinal ballast
resistance builds up progressively from the ends of the long
welded rail towards the centre.
If the track frame was not restrained, then the rail would expand
or contract with variation in temperature and consequently no force
would build up in the rail. However, since there is a restraint now
offered by the longitudinal ballast resistance, forces are set up in the
rail. These are called thermal forces. If the temperature variation
(from the temperature at which the rail was fastenced to the sleeper)
is small, then a small length of track at the end would be sufficient to
develop longitudinal ballast resistance against the tendency for free
movement of the rail. Consequently, it would lead to a small amount
of thermal force in the rails. However, If the temperature difference
becomes more, a longer length of track at the ends would be called
upon to develop the necessary longitudinal ballast resistance
against the free movement of the ends of the rails and there would
be a corresponding increase in the thermal forces.
There is however, a limit up to which the temperature
differences can build up. This limit is dictated by:
a) either the maximum or the minimum rail
temperature and
b) the temperature at which the rail was fastened
to the sleeper.
The temperature at which the track could be attended under
regular track maintenance, the temperature at which the patrolling
of the track should be introduced, etc. are governed by the sole
consideration that the thermal forces at any time in the track
should be within safe limits, to avoid the eventualities of buckle
and fracture of the LWR.
(10)
2.2 Force Diagram
In the previous paragraph it has been indicated that thermal
forces are built up in the LWR due to the resistance offered by the
ballast to the free expansion or contraction of the track frame. An
expression for the magnitude of this thermal force is required as
this will govern the design of the LWR. Let us assume the LWR
central portion to undergo a change of temperature equal t
o
C
temperature rise. If the central portion of length ‘L' had been free
to expand it would have expanded by an amount equal to ‘Lαt’.
However since the central portion of the LWR does not move, the
compressive strain induced in central portion is equal to
t
L
t L
α
α
·
where ‘t’ is the change of temperature of the LWR
with respect to the temp. at which the LWR was laid or
destressed and α is the coefficient of linear expansion.
If P is the force induced in central portion (compressive
force) and A is the cross section area of rail, then P/A is the
compressive stress in the rail.
Since stress /strain = E (Young’s modulus of rail steel)
we have
E
t
A P
·
α
/
or P = A E α t where P is in Newtons, A is in mm
2
, and E is in N/
mm
2
.
As t represents the change of temperature of the rail w.r.t. the
temperature it is stress free, a technically correct formula for the
thermal force in the LWR will be
P = A E α (t
P
-t
n
)
where t
P
is the prevailing rail temperature and t
n
is the rail neutral
temperature, which is the temperature at which the LWR is free of
longitudinal thermal stress.
(11)
(12)
When t
P
= t
n
, P=0,
when t
P
>t
n
, P is a compressive force
and when t
P
< t
n
, P is a tensile force.
As t
n
is not known it is assumed that t
n
= t
L
or t
d
, temperatures at
which LWR was laid (t
L
)or destressed (t
d
).
The force at the beginning of the LWR is zero, and in the
central portion equal to A E α t. This change of force from zero to
a peak value occurs over the breathing length. The shape of the
force diagram is therefore as given in Figure 2.1
Thermal stress in rail due to unit degree change of temperature
= E x α = 2.15 x 10
6
X 1.152X10
-5
= 2.5 MPa/
0
C
For 30
0
C change of temperature therefore thermal stress induced
will be about 75 MPa which is about 8.5% of the ultimate tensile
strength of a 90 UTS rail.
2.3 Importance of Rail Temperature
From the expression for force in an LWR, P = AEα(t
P
- t
n
), it
can be seen that force developed in the LWR depends primarily on
the prevailing rail temperature and the rail neutral temperature.
Both these factors are discussed below:
Fig. 2.1
2.3.1 Rail Temperature Measurement :
Thermometers
The following are the different types of approved
thermometers for measuring rail temperature :
i) Embedded type - This is an ordinary themometer
inserted in a cavity formed in a piece of rail-head, the
cavity filled with mercury and sealed. The rail piece is
mounted on a wooden board which is placed on the
cess and exposed to the same conditions as the rail
inside the track. This type of thermometer takes 25 to
30 minutes for attaining temperature of the rail.
ii) Dial type - This is a bi-metallic type thermometer
which is provided with a magnet for attaching it to the
rail. The thermometer is attached on the shady side of
the rail web as this location is approximating the
average rail temperature to the greatest extent. A
steady recording of the rail temperature is reached
within 8 minutes.
iii) Continuous recording type - It consists of a graduated
chart mounted on a disc which gets rotated by a
winding mechanism at a constant speed to complete
one revolution in 24 hours or 7 days as applicable,
giving a continuous record of rail temperature. The
sensing element is attached to the web of the rail and
connected to the recording pen, through a capillary
tube wihch is filled with mercury.
iv) Any other type of thermometer approved by RDSO/
Chief Engineer.
Where a number of thermometers are used to
measure the rail temperature at one place, as in case
of laying of LWR, destressing etc. any of the
thermometers showing erratic readings, appreciably
different from the other adjoining thermometers, shall
be considered as defective.
(13)
Zonal Railways should nominate 8 to 10 stations in their railway
in a manner as to give the representative sample of the
temperature variations on the Zonal Railway for the region allotted
to each station. These stations shall be the existing PWIs offices.
On these stations rail temperature records shall be built up using
preferably a well calibrated continuous recording type thermometer.
The maximum and minumum rail temperature for a continuous
period of at least 5 years shall be ascertained and the mean rail
temperature (t
m
) for the region arrived at. These temperature
records shall be analysed to assess the probable availability of
time periods during different seasons of the year for track
maintenance, destressing operations and requirements of hot/ cold
weather patrolling etc. Rail thermometers shall also be available
with each each gang and sectional PWIs to enable the gangs to
work within the prescribed temperature ranges.
2.3.2 Rail Temperature Zones and RDSO Studies :
In order to understand the correlation between the rail
temperature and ambient temperature, RDSO conducted rail
temperature studies between 1969 and 1971 over a two year
period. 22 stations were identified over the Indian Railways where
Standard Measuring Arrangements for Rail Temp (SMART) were
set up. (Fig. 2.2) SMART consisted of a full-length rail laid in the
east-west direction on wooden sleepers with ACB plates and boxed
with standard ballast profile. The rail temperature was measured by
means of a thermometer placed in a mercury-filled hole in the rail
head. Rail temperature readings were taken on an hourly basis
between August 1969 and August 1971, and the corresponding air
temperatures obtained from the Meteorological office. Correlation
equations between the rail temp and air temp were derived using a
computer based regression analysis (Details available in RDSO/C/
146, Reference-3) for each of the identified 22 stations. Using these
correlation equations and the maximum and minimum air
temperatures at 180 stations over the Indian Railways, obtained
from the Weather office over a period of 90 years, it was possible to
determine the maximum and minimum rail temperatures obtainable
at these stations. This data was presented in the form of a rail
(14)
Fig. 2.2 : STANDARD MEASURING ARRANGEMENT
FOR RAIL TEMPERATURE.
(15)
temperature map where against each station two figures are
given, namely the rail temperature range and the rail temperature
mean.
The Rail Temperature Range = Max Rail Temp - Min Rail Temp
and Mean Rail Temperature = Max Rail Temp + Min Rail Temp
2
The country is further divided into four zones depending
upon the temp range :
Temp. Zone Temp. Range
I 40-50
o
C
II 51-60
o
C
III 61-70
o
C
IV 71-76
o
C
The rail temperature map is given in the LWR Manual (1996).
2.3.3 Rationale behind choice of t
n
or t
d
The LWR neutral temperature should be chosen in such a
manner that the thermal force developed in the LWR is within the
desired limits.
Refer to Fig 2.3. The block shows the maximum, minimum
and range of rail temperature in the four temperature zones. The
rail can be fixed to the sleepers by fastenings after destressing, at
a temperature anywhere within the range between maximum and
minimum rail temperatures.
Let us see what happens if we fasten the rail to the sleeper at
the minimum rail temperature(t
min
). As the rail temperature rises
compressive thermal forces will be built up and when the rail
temperature reaches t
max
compressive forces proportional to the full
range of rail temperature will be built up. Such large compressive
forces could be very dangerous to the stability of the LWR and the
track can buckle. In this case there is of course no danger of any
tensile force developing in the rail and consequently of rail fracture.
(16)
Fig. 2.3 : VARIATION OF TEMPERATURE WITH RESPECT
TO t
d
IN DIFFERENT ZONES.
t
min
t
min
t
min
t
min
(17)
Let us see what happens if the rail is fastened to the sleeper
at the maximum rail temperature (t
max
). Since temperature cannot
rise any further, there is no likelihood of compressive thermal
stresses developing in the rail, and consequently there is no
danger of buckling. However, since the rail temperature can fall
through the complete range of temperature at this place, tensile
stresses and forces in the rail could develop to a very large
magnitude making rail fracture very probable.
Logic suggests that we should fix the destressing
temperature exactly mid-way between maximum and minimum rail
temperatures. i.e. at mean rail temperature. In that case the extent
of maximum compressive or maximum tensile forces would be
equal and half of what it would otherwise be as in case of either of
the two previous situations. All the same compressive forces and
tensile forces are bound to develop and so the possiility of buckling
as well as fracture will exist. A fracture will create a gap in the rail.
However, in the case of fractures, the alignment of the rail is not
immediately distorted. Also fractures rarely occur on both rails and
at the same location simultaneously. Thus at least a few trains can
pass over a fractured rail without accident till the fracture is
attended to. However, if the track buckles due to excessive
compressive forces in the rail, alignment of track gets distorted and
safe running of trains is endangered.
Therefore considering buckling more dangerous, it is
considered prudent to fix the destressing temperature higher than
the mean rail temperature so that the compressive forces built up
in the track would be within reasonable limits, though at the cost of
introducing higher tensile forces. This is the basis for fixing t
d
on
the lndian Railways at a temperature above t
m
. Since 90R and
lighter sections do not have adequate margin to accommodate the
resulting thermal tensile stresses, t
d
for such rails has been fixed
between tm and t
m
+ 5
0
C. Heavier 52 kg and 60 kg rails having
greater section modulus can accommodate relatively larger thermal
tensile stresses and so t
d
for these rails has been fixed between t
m
+ 5 and t
m
+ 10
0
C. It also gives relief by reducing the compressive
forces which are directly proportional to area of cross section of
the rail, other conditions remaining the same.
(18)
Since the operation of fastening of the rails to the sleepers
after destressing takes time during which rail temperature can vary,
a range has been recommended for t
d
instead of a fixed value.
In addition to the above consideration of avoiding buckling
while risking the chances of fracture there are some more reasons
why the destressing temperature has been fixed at a level higher
than t
m
.
1) Studies on LWR behaviour have indicated that the
stability of the track gets endangered at
temperatures above t
d
+ 10
0
C. Hence the clause of
the LWR Manual limiting the maintenance
operations on the LWR to a rail temp of t
d
+ 10
0
C
has been provided. If the destressing temperature
of the rail is lowered the prevailing rail temperature
could rise beyond t
d
+10
0
C more frequently
reducing the availability of maintenance hours. This
would either entail limited working hours to the
gangs, night working or working in split shifts.
These options have serious practicability problems.
Hence the destressing temperature of the LWR has
been fixed at a temperature above the mean temp
so that the prevailing rail temp does not frequently
go above t
d
+ 10
0
C, restricting the maintenance
operations, and ensuring adequate maintenance
hours.
2) The other provision of the manual is that hot weather
patrolling should be introduced when t
P
rises above
t
d
+ 20
0
C. Again if t
d
is set at a lower temp. as
compared to t
m
, the prevailing rail temperature could
rise above t
d
+ 20
0
C quite often leading to an
increase in the quantum of hot weather patrolling.
3) The LWR manual prescribes imposition of speed
restrictions at locations where track maintenance
activities have been carried out and the temperature
rises beyond t
d
+ 20
0
C during the period of
consolidation. If t
d
is fixed at a lower level, this would
necessitate imposition of a large number of
(19)
restrictions at work sites, whenever the rail
temperature rises beyond t
d
+20
0
C during the
period of consolidation. This situation is
unacceptable.
Keeping the above factors in mind t
d
has been fixed
at a higher level as compared to t
m
.The policy
makers at that time knew that doing this will have
an effect on the number of fractures and therefore
indicated that a greater frequency of USFD
inspection and monitoring would be required.
Fractures are today one of our major concerns and
there is a demand from some railways to reduce
the level at which the destressing temperature
should be fixed basically to reduce the number of
fractures. A more realistic view would be to have a
flexible neutral temperature which could be shifted
downwards in winter to control fractures and shifted
upwards in summer to control the tendency of the
track to buckle.
2.4 Breathing Length
2.4.1 As discussed earlier the force build up in the LWR starts in
the breathing length at the free end. The tendency of the rail
sleeper assembly to expand or contract is resisted by the
resistance of the ballast called the longitudinal ballast resistance. It
is denoted by R and its unit of measurement is in kg/metre/rail.
The longitudinal ballast resistance is mobilized when there is a
relative movement between the sleeper and the ballast. If LB is
breathing length and R the longitudinal ballast resistance, then LB
x R represents the total resistance offered by the ballast in the
breathing length.
As the maximum force in the LWR = AE α t
LB x R = AE α t or LB =
R
t AEα
The above expression for the breathing length indicates the
various factors which govern the breathing length as under:-
1) The breathing length is proportional to the temperature
(20)
change. Therefore the breathing length is maximum in
Zone IV and minimum in Zone I.
2) The larger the cross sectional area of the rail, the
larger the breathing length.
3) The larger the value of 'R' the longitudinal ballast
resistance, the smaller the breathing length. As BG
sleepers have a larger value of R the breathing length
of BG LWRs is smaller as compared to MG LWRs.
Annexures 1A, 1B & 1C of LWR Manual give indicative values of
breathing lengths for different sleepers and sleeper densities in
different zones.
2.5 Longitndinal Ballast Resistance
The longitudinal ballast resistance ‘R’ comes into play when there
is relative motion of the sleepers with respect to the ballast in the
longitudinal direction.
2.5.1 RDSO conducted a number of experimental studies on the
various aspects of the longitudinal ballast resistance. These
studies are described in RDSO Report No. C-148(Reference 4).
Experimental Set up : These studies were conducted
experimentally on a running track as well as on a freshly laid track
in the lab. A track section comprising of short length rails and
three sleepers embedded in ballast was pushed in the longitudinal
direction and the displacement versus load curve plotted. The
tests were conducted on a running line where a traffic block of 90
minutes was taken. The load was applied to this test panel by a
hydraulic jack with a remote controlled pumping unit. The load
applied was measured with the help of a proving ring and the
longitudinal movement of the panel was recorded with the help of
dial gauges. The instantaneous loads and movements were
measured as the load was increased gradually till it reached a
peak and fairly steady value. After the test, the short length rails
were replaced by the normal rails. (Refer to figure 2.4)
The maximum value obtained from the proving ring was divided by
twice the number of sleepers in action to get the ballast resistance
(21)
Fig 2.4
(22)
per sleeper per rail. This figure divided by the sleeper spacing in
metres gives the value of 'R' in kg/m/rail.
Findings of the study :
1. In BG, PRC sleepers give the highest longitudinal
ballast resistance in all conditions. In consolidated and
through packed conditions other sleepers in order of
decreasing longitudinal resistance are: Steel, RCC
twin block, CST-9 and Wooden. In deep screened or
freshly laid track, the order is PRC, RCC, Steel,
Wooden and CST-9.
2. Through packing causes a reduction in ballast
resistance. For both BG & MG concrete sleepers the
average reduction is 23%. For BG conventional
sleepers the reduction is 36%.
3. Deep screening causes a greater reduction in ballast
resistance.
4. The effect of traffic on the growth of the ballast
resistance is substantial. BG concrete sleepers attain
86% of the consolidated value on passage of 1 GMT
of traffic.
5. Ballast resistance per sleeper decreases as the
sleeper spacing reduces, but the ballast resistance per
unit length of track remains more or less constant for
sleeper densities from 1200 nos. to 1500 nos. per km.
For larger sleeper densities the value of the
longitudinal ballast resistance again increases due to
heavier track structure.
6. A heaped up shoulder ballast gives a higher ballast
resistance as compared to the standard shoulder for
both BG and MG. This increase is maximum for
concrete sleepers.
A summary of values obtained for different sleepers under
different conditions is given in Table1, Table2 and Table3.
(23)
(24)
Table 1
Longitudinal Ballast Resistance (kg/m/rail)
(Effect of sleeper type and track maintenance activities)
Gauge Sleeper Conso- Through % Deep %
Type lidated packed Loss Screened Loss
BG PRC 1244 1027 17 885 29
Steel 1051 744 29 433 59
CST-9 933 551 41 276 71
RCC 921 666 28 581 37
Wooden 697 380*/552 45*/21 370 47
MG Wooden 405 180*/302 55*/21 259 36
Steel 298 231 22 209 30
* Values are for tracks packed by MSP
Conventional sleepers are beater packed
PRC sleepers have been off track tamped.
Sleeper spacing PRC – 1600 Nos/km
RCC and conventional – 1200 Nos/km
Adapted from RDSO/C-148
Table 2
Maximum Longitudinal Ballast Resistance (kg/m/rail)
in freshly laid conditions
(Effect of sleeper density & heaped up ballast profile)
38 mm Standard Ballast 38 mm Heaped up Ballast
Ballast profile profile
Sleeper A B C D A B C D
PRC 546 559 561 628 — — 648 —
Steel 347 375 387 434 — — 438 —
A : 1200 Nos per km B : 1400 Nos per km
C : 1600 Nos per km D : 1800 Nos per km
Adapted from RDSO/C-148
Table 3
Longitudinal Ballast Resistance (kg/m/rail)
(Effect of movement of traffic)
Sleeper consolidated Resistance % of
Value on passage consolidated
of 1 GMT of Value
traffic
PRC 1244 1072 86
Steel 1051 740 70
CST- 9 933 580 62
RCC 921 750 81
Adapted from RDSO/C-148
(25)
2.6 Lateral Ballast Resistance
The lateral ballast resistance comes into play when the track
has a tendency to get displaced in the lateral direction due to build
up of compressive forces. RDSO studies(Reference 5) conducted
on various aspects of lateral ballast resistance have indicated the
values of lateral ballast resistance as given in the Table below. The
test setup is given in Fig. 2.5.
Values of Lateral Ballast Resistance in Kg/m of track
Gauge : BG Conso- Through Deep
Sleeper lidated Packed Screened
CST-9 1640 1100 532
PRC 1470 1226 1040
Steel 1430 825 540
RCC 1420 1040 800
Wooden 1060 320/520 384
Adapted from RDSO/C-156
1. The higher resistance recorded by metal sleepers is
due to the central keel in the case of CST-9 sleepers
and turned down ends in case of steel sleepers which
get embedded in the ballast core and offer better
resistance to the lateral movement.
2. The reduction in the lateral ballast resistance on
through packing and deep screening is quite
significant for CST-9 and steel sleepers and much less
for PRC sleepers.
Tests conducted have revealed that :
u Track surfacing and ballast tamping even with a minor
amount of rail lift (0.5 to 1 inch) can cause significant
reduction in lateral track strength.
(26)
u Depending upon the ballast type, recovery of strength
loss due to traffic could vary from 0.3 GMT to 9 GMT.
u Dynamic track stabilizers could significantly acclerate
ballast consolidation or strength recovery. For
instance. for granite ballast, the dynamic track
stabilizer may produce a consolidation equivalent to
0.3 GMT.
(27)
f Fig. 2.5
Dial gauge
CHAPTER III
THERMAL MOVEMENTS AND HYSTERESIS
3.1 Estimation of Thermal Movements
It is only in the breathing lengths that a LWR displays the property
of longitudinal movements. At the ends of the LWR since the restraint
offered by the longitudinal ballast resistance is nil, the movement is
observed to be the maximum. As the longitudinal ballast resistance
exerted on the sleepers progressively builds up complimentary forces
in the rail increase from A towards B.(Fig.3.1) At B, which is the junction
between the breathing and fixed lengths, the movement reduces to
zero. The movements recorded in the field at various points in the
breathing length of the LWR corroborate the above mentioned
observations. It is possible to make certain simplifying assumptions
and estimate the movement at any point in the breathing length.
Fig. 3.1
P(x)
(28)
Take a small length of rail dx at any arbitrary point M at a distance
‘x’ away from B (refer fig. 3.1).
It is possible to calculate the amount of free expansion of the small
rail of length ‘dx’ due to change of rail temperature as well as the amount
of contraction due to presence of thermal force present in the rail at
that length:
Free expansion of this small length dx due to a rise in rail
temperature by t
0
C = dxα t. The amount of contraction of this length
dx, is equal to P(x) dx
E A
(where P(x) is the thermal force present in the small length of rail dx at
a distance x away from B).
The net expansion of the small length of rail dx will therefore be the
difference between the above two values. If this net expansion is called
dy, then
AE
dx x P
tdx dy
) (
− ·α
dx
AE
x P P
AE
dx x P tdx AE
1
]
1

¸

·

·
) ( ) ( α
where P= AE α t is the maximum thermal force in the central portion
of the LWR.
Integrating the net expansion of all such small lengths of rails starting
from B towards M, we can obtain the total expansion or displacement
of points M as

· ·
x
AE
dy Y
0
1
[ ]


x
dx Px P
0
It can be observed from Fig. 3.1 that the expression (P-P(x))dx is
nothing but the area of the shaded diagram appearing above the
diagram of thermal force.
Thus the amount of maximum contraction or expansion at any point
( )
(29)
in the breathing length of a LWR can be computed by dividing the
shaded area from B, upto that point as in the Fig. 3.1 by EA. Extending
this logic, the cumulative value of expansion or contraction at the end
of the LWR i.e. at ‘A’ or ‘D’ can be obtained as follows:
Maximum expansion or contraction at ‘A’ or ‘D’ =
AE
FE ngleA AreaofTria
1
= 1/2 x P(LB)/AE .... (Equation 1)
As P= AE α t, equation (1) above can be simplified as maximum
movement at the end of a LWR, m
= ((LB) α t)/2 .... (Equation 2)
We can look at it this way :
The maximum movement of the end of the LWR is half the
corresponding value if only the breathing length LB of the LWR is
allowed to expand or contract absolutely freely. The variation of
movement along the breathing length is given in Fig.3.2.
This expression for the maximum movement at the end of the LWR
i.e. m = (LB α t )/2 can also be rewritten as:
(30)
l l
l
l
2
t LBα
Breathing
Length
f f
Fig. 3.2
Variation of movement in Breathing length
Central portion
m · ΑΕ (α t)
2
/2R ... (Equation 3)
This can be derived from Equation 1 by substituting AEα t/R in
place of LB because
LB = P/R = AE α t/R
It should be noted that in the above calculations an important
assumption has been made that in a breathing length of LWR, the
sleepers have equal values of longitudinal resistance.
To illustrate the above let us solve an example with the data given
below:
Gauge: BG, Sleepers: PRC, Rails: 52kg (A=66.15cm
2
)
Sleeper density 1540 sleepers/km, Zone IV with temp. range
=76
0
C
R = 13.28 kg/cm/rail
A = 66.15 cm
2
, E = 2.15 x 10
6
kg/cm
2
t
d
= t
m
+ 10
0
C
t = 28
0
C for temp. rise
t = 48
0
C for temp. fall
α= 1.152 x 10
-5
/
0
C
As discussed earlier movement at an SEJ
R
t AE
2
) (
2
α
· ∆
For a temperature rise of 28
0
C,
( )
28 . 13 2
28 10 152 . 1 10 15 . 2 15 . 66
2
5 6
1
×
× × × ×
· ∆

= 5.57 mm (forward movement of tip of tongue/stock rail wrt
reference line)
(31)
For a temperature fall of 48
0
C
( )
28 . 13 2
48 10 152 . 1 10 15 . 2 15 . 66
2
5 6
2
×
× × × ×
· ∆

= 16.37 mm (backward movement of tip of tongue/stock rail wrt
reference line)
Total movement at the SEJ joint
= 2 x (5.57 + 16.37)
= 2 x 21.94
= 43.88 mm
The movements which occur at the SEJ due to thermal variations are
shown in Fig. 3.3 (A,B,C) with values as calculated in the above
example.
At tp=td
At tp=td+28
0
C
At tp=td-48
0
C
Fig. 3.3
(32)
3.2 Switch Expansion Joints (SEJ) :
The thermal movement in the breathing length of an LWR are
accomodated at the switch expansion joint (SEJ). An SEJ typically
consists of a pair of tongue rails and stock rails , the tongue rail laid
facing the direction of traffic. Modern SEJs are laid on concrete sleepers
with rail free fastenings. The tongue rails and stock rails are machined
and given suitable bends to accomodate each other. Tongue rails and
stock rails have typically straight alignment and hence these SEJs
cannot be laid in curves sharper than 0.5
0
. The distance between the
tip of the tongue rail and notch of the stock rail is typically kept as
40mm at the destressing temperature. Earlier the gap used to be 60mm
and has now been reduced to 40mm to lower the impact of the passing
wheel. Various types of SEJs used on Indian Railways are described
below:
Drawing No. Description
RDSO/T-4160 Assembly for Switch Expansion Joint with 80 mm
max gap for LWR BG 52 kg on Concrete sleepers
RDSO/T-4165 Assembly for SEJ with 80 mm max gap for LWR
BG 60 kg on concrete sleepers
RDSO/T-5748 Assembly for SEJ with LWR BG 60 kg (UIC) on PSC
sleepers laid on curve with curvature from 0.5
0
to 1.5
0
RDSO/T-6039 Assembly for SEJ with 190 mm max gap for bridge
approaches for LWR BG 52 kg on concrete sleepers
RDSO/T-6263 Assembly for SEJ with 190mm max gap for bridge
bridge approaches for BG 60 kg(UIC) on PSC
sleepers
RDSO/T-6257 Assembly for SEJ for 80 mm max gap with BG
CR – 120 crane rails on PSC sleepers
(i) RDSO/T- 4160 and RDSO/T- 4165 are the conventional straight
SEJs with 80 mm maximum gap. Each SEJ has a pair of tongue rails
and stock rails, with 6 special sleepers to RDSO drawing No.RDSO/T-
4149. All these are 300 mmwide sleepers with sleepers Nos 10 and 11
(33)
F
i
g
.


3
.
4
A
(34)
T
y
p
i
c
a
l

S
E
J

l
a
y
o
u
t
A
n
g
l
e

t
i
e
s
l l
(35)
F
i
g
.


3
.
4

(
a
)
R
A
I
L

R
E
F
E
R
E
N
C
E

P
O
S
T
with special fastenings and sleeper No. 8, 9, 12 and 13 with similar
fittings. The centre line of sleeper No. 10 coincides with the tip of the
tongue rail and the 40 mm initial gap is provided with the tip of the
tongue rail coinciding with the centre of the sleeper No. 10. The centre
to centre spacing of sleeper No. 10 and 11 is 700 mm while the sleepers
spacing from 1 to 10 and 11 to 20 may be 600 mm or 650 mm depending
upon the sleeper density. Fig. 3.4 gives details of a typical SEJ layout
and 3.4(a) gives the details of location A.
(ii) RDSO/T-6039 & RDSO/T-6263 : These are wide gap SEJs for
bridge approaches where the maximum gap permitted is 190 mm, the
mean position is kept at 166 mm from centerline of sleeper No. 10 to
enable the tongue rail to remain on the sleeper even when the entire
expansion takes place. Sleepers No. 7, 8, 9, 10, 11, 12, 13 are special
sleepers with sleeper nos. 9, 10 and 11 with special fastenings. Use is
made of ERC Mark II clips with flat toe designed with a toe load of 350-
400 kg/clip to enable free rail movement. Sleepers other than 7 to 13
are approach concrete sleepers with normal fittings. When gap is more
than 100mm for passing diplorries with smaller wheel diameter, use of
an insertion piece in the gap should be made.
(36)
Fig 3.5
curvature > 0.5
o
& < 1
o
curvature > 1
o
& < 1.5
o
p p
11690 mm
p p
11690 mm
(iii) RDSO/T-5748 : These SEJ layouts can be used when the SEJ has
to be laid in a curve sharper than 0.5
0
but not sharper than 1.5
0
. The
tongue rail and stock rail are given curvature as given (Fig. 3.5).
The conventional SEJ design involves two bends in the stock rail and
tongue rail which are locations of weakness resulting in fractures.
Improved design SEJs developed by various industries are under trial
on the Railways till final approval is received from the Railway Board.
A brief description of these layouts is given below:
(1) SEJ with one gap : This design has been developed by M/s Rahee
Industries Ltd, Calcutta. Fig 3.6 The design comprises of a pair of
machined segments on the non-gauge face side of two non-bent
running rails mounted with a gap between the juxtaposed rail ends
and a third rail called a gap avoiding rail of predetermined length
accommodated in the said machined segments parallel to and adjacent
to the non bent straight length of the running rails. This rail is securely
fitted to one of running rails with high tensile steel bolts. This running
rail together with the gap avoiding rail is called the stock rail. The other
running rail is called the tongue rail. The non-bolted segment of the
gap avoiding rail braces the machined segment of the tongue rail.
11 1
HTS
BOLTS
Fig 3.6 ONE GAP SEJ
(37)
Gauge Face Side →
Advantages :
1) No bends in tongue and stock rail.
2) Only 5 special sleepers of standard SEJ on PSC assembly
are used.
3) Check rails guard against excessive play of worn out
wheels.
4) Design suitable upto 180 mm maximum gap.
(2) SEJ with two gaps
Two designs have been developed by two different firms (M/s Bina
Metal way, Jamshedpur and M/s Chi ntpurni Engi neeri ng
Works,Barabanki). In both these designs two gaps of maximum 80
mm each are provided in one SEJ. Thus a maximum gap of 80 mm is
available for an LWR on one side of the SEJ. Similarly a gap of 80mm
is available for the LWR on the other side. The tongue rail is
manufactured by cutting the rail at head and foot location. Two cut
rails are joined together to make the stock rail.
Salient features of Bina Metalway 2-gap SEJ (Fig 3.7)
The stock rail is considered to be static with negligible expansion and
contraction in length due to temperature changes. This SEJ makes
use of 6 wider concrete sleepers each to Drg No. T/4149, with three
sleepers located near each gap. The length of the SEJ is 5750 + 6950
+ 5920 + 80 = 18700 mm.
Hence a total gap of 18750 mm should be created while inserting this
SEJ. The stock rail is fabricated out of two pieces of lengths 7140 mm
and 5920 mm connected to each other by HTS bolts. While laying the
SEJ it should be ensured that the ends of the stock rail are 40 mm
away from the centre line of sleeper Nos. 12 and 22 with the tip of the
tongue rail coinciding with the centre line of the sleeper.
1. Sleeper Nos. 1 to 31 should be at a spacing of 600 mm c/c.
2. Sleeper Nos. 10, 11, 12, 22, 23 and 24 are special sleepers to
RDSO drawing No. T – 4149 and the rest are normal PSC line
sleepers.
(38)
FIG 3.7 TWO GAP SEJ
stock rail
Tongue rail
Tongue rail
(39)
3. Mean position of SEJ should be kept at centre line of sleepers
No. 12 and 22.
4. The mean gap is 40 mm on each end.
5. The tongue rails are kept at mean position at centre line of
sleeper Nos. 12 and 22, and stock rail end kept at 40 mm from
mean position, thus creating a gap of 40 mm.
6. The mean position should also be marked on the rail posts
erected on both sides of track.
3.3 Phenomenon of Hysteresis :
The behaviour of an LWR, as far as movement at the SEJ is
concerned, is irrational which will be evident form the (Fig. 3.8) shown
below :
As the temp. uniformly rises above ‘O’, the movement or expansion
at the SEJ follows the movement – temp. rise curve OF where
) )
) )
) )
) )
Fig 3.8 HYSTERESIS PHENOMENA
C
B
E
D
(40)
F
A1
( )
R
t AE
2
2
α
· ∆
At any given temperature t
4
if the temp starts falling, then the movement
at the SEJ does not follow the original path but traces out a new curve
A
2
DB
2
. If at B
2
the temp again reverses then the path traced out is
B
2
EA
2
rather than B
2
DA
2
Loops in the form A
2
DB
2
EA
2
are called
hysteresis loops and are formed whenever there is a temperature
reversal. In order to simplify matters, an annual hysteresis loop or curve
is plotted which will envelope all the hysteresis loops formed on a daily
basis. (Fig. 3.9)
This hysteresis loop can be traced in two ways :
1) Temp rise from t
d
to t
max
, fall from t
max
to t
min
and again a rise to
t
max
from t
min.
2) Temp fall from t
d
to t
min
, rise from t
min
to t
max
and then a fall from
t
max
to t
min
.
While plotting these curves it should be remembered that whenever
there is a reversal of temperature, the longitudinal ballast resistance
should be taken as twice its normal value ( 2 R instead of R). The final
hystersis loop shall be an envelope of the two hysteresis loops as
drawn above.
Implications of hysteresis:
At temp. t
P
the movement at the SEJ may be an expansion equal to ‘a’
or a contraction ‘b’. This would mean that the gap at the SEJ could be
(20 – a) or( 20 + b) if 20 mm is the initial gap. Hence, due to hysteresis
the gap at SEJ is not a discrete value but a range. LWR Manual
Annexure V gives the permissible range of gaps at the SEJ for different
track structures at different rail temp. for different zones as given on
the adjoining page :
(41)
ANNUAL HYSTERESIS LOOP
SHOWING THE MOVEMENTS OF ENDS OF L.W.R. PROVIDED
WITH SWITCH EXPANSION JOINTS
Fig. 3.9
(42)
Temperature Zone I Zone II Zone III Zone IV
t
d
+ 28 – – – 14
t
d
+ 25 – – 16 14 – 16
t
d
+ 20 – 17 16 to 18 15 – 19
t
d
+ 15 18 17 to 19 16 to 20 15 – 21
t
d
+ 10 18 to 20 18 to 21 16 to 22 16 – 23
t
d
+ 5 19 to 22 18 to 23 17 to 24 16 – 25
t
d
19 to 23 19 to 24 18 to 26 17 – 27
t
d
– 5 20 to 24 19 to 26 19 to 27 18 – 28
t
d
– 10 21 to 25 20 to 27 20 to 29 20 – 30
t
d
– 15 22 to 26 22 to 28 21 to 30 21 – 31
t
d
– 20 23 to 26 23 to 29 23 to 31 23 – 32
t
d
– 25 24 to 27 24 to 29 24 to 32 24 – 33
t
d
– 30 26 to 27 26 to 30 26 to 32 26 – 34
t
d
– 35 27 28 to 30 28 to 33 29 – 34
t
d
– 40 -- 30 31 to 33 31 – 35
t
d
– 45 -- -- 33 33 – 35
t
d
– 48 -- -- -- 35
SAMPLE TABLE FROM ANNEXURE V OF LWR MANUAL
GAPS BETWEEN REFERENCE MARK AND TONGUE RAIL TIP/
STOCK RAIL CORNER OF SEJ FOR VARIOUS TEMPERATURES
IN mm FOR BG, 52 Kg, PRC SLEEPER, 1540 Nos/km, VALUE OF
R (BALLAST RESISTANCE) ASSUMED = 13.28 Kg/cm/rail AND t
d
AS PER PARA 1.11
(43)
The gaps are with an initial setting of 40 mm. When gaps have been
initially laid at 60 mm, 10 mm should be added to each of these values.
The gaps given in the above table are the theoretical gaps. A tolerance
of + 10mm is prescribed beyond the given gaps.
Cause of Hysteresis : Hysteresis is due to the behaviour of the
longitudinal ballast resistance. A plot of the resistance offered by the
ballast vis-à-vis the sleeper displacement is as given in Fig 3.10.
The ballast resistance first increases linearly as the sleeper
displacement, then goes into the plastic zone and finally assumes a
constant value R. If at this stage the temperature reverses then the
value of the longitudinal ballast resistance drops to zero and then
becomes (-R) as shown above. This shows that at the time of reversal
of temp the ballast resistance mobilized is 2R. Due to this effect, the
path traced out at the end of LWR follows an irregular path leading to
the hysteresis phenomenon.
Fig. 3.10
(44)
‘X’
G F -R
‘Y’
+R



3.4 Gap measurements at an SEJ :
At the SEJ a reference line is established between the tongue rail and
stock rail. This gap between the tongue rail and stock rail will be equal
to 40mm for 52kg and 60kg rail sections, and for other rail sections
60mm as shown. (Fig. 3.11 )
Gaps g1 and g2 are not discrete values but the permissible range has
been defined in the LWR Manual Annexure V for different track
structures, different zones and different prevailing temperatures. A
sample page for filling up the movements observed at an SEJ as per
annexure XIII A of the LWR Manual is shown in Fig. 3.12.
GAP MEASUREMENT AT AN SEJ
Fig. 3.11
(45)
A
n
n
e
x
u
r
e

X
I
I
I

A
C
h
a
r
t

o
f

m
o
v
e
m
e
n
t

o
f

L
W
R
/
C
W
R

N
o
.
1
0
S
E
J
s

a
t

t
h
e

e
n
d

o
f

t
h
i
s

L
W
R

:

S
E
J

N
O
.
1
0

a
t

k
m

;

S
E
J

N
o
.
1
1

a
t

k
m

D
a
t
e

o
f
T
i
m
e

o
f
R
a
i
l
R
i
g
h
t

o
r
D
i
s
t
a
n
c
e
(
m
m
)
D
i
s
t
a
n
c
e

(
m
m
)
c
/
c

s
p
a
c
e
M
e
a
s
u
R
e
c
t
i
f
i
R
e
m
a
r
k
s
M
e
a
s
u
r
e
M
e
a
s
u
r
e
T
e
m
p
L
e
f
t

R
a
i
l

b
e
t
w
e
e
n
t
o
n
g
u
e
/
b
e
t
w
e
e
n

t
o
n
g
u
e
/
b
e
t
w
e
e
n
r
e
d

b
y
c
a
t
i
o
n
m
e
n
t
m
e
n
t
s
t
o
c
k

r
a
i
l

&
s
t
o
c
k

r
a
i
l

&
s
l
e
e
p
e
r
s

t
h
e

t
w
o
c
a
r
r
i
e
d
R
e
f
e
r
e
n
c
e

l
i
n
e
R
e
f
e
r
e
n
c
e

l
i
n
e
c
e
n
t
r
a
l

s
l
e
e
p
e
r
s
o
u
t
a
t

S
E
J

N
o
.
1
0
a
t

S
E
J

N
o
.
1
1
O
b
s
e
r
P
e
r
m
i
s
O
b
s
e
r
v
P
e
r
m
i
s
a
t

S
E
J
a
t

S
E
J
o
n

b
y
v
e
d
s
i
b
l
e
e
d
s
i
b
l
e
N
o
.

1
0
N
o
.
1
1
d
a
t
e
(
a
)
R
a
n
g
e
R
a
n
g
e
6
.
1
0
.
2
0
0
4
1
0
:
0
0
3
0
0
C
L
e
f
t

r
a
i
l
g
2
R
i
g
h
t

r
a
i
l
g
4
6
.
1
0
.
2
0
0
4
1
4
:
0
0
4
0
0
C
L
e
f
t

r
a
i
l
g
5
R
i
g
h
t

r
a
i
l
g
7
F
i
g

3
.
1
2
-

S
a
m
p
l
e

o
f

S
E
J

g
a
p

m
e
a
s
u
r
m
e
n
t
s
.

A
l
s
o

r
e
f
e
r

F
i
g

3
.
1
1
(46)
CHAPTER IV
PERMITTED LOCATIONS AND
TRACK STRUCTURE
4.1. General Considerations :
1. It is a policy that all new constructions, gauge conversions,
doublings and permanent diversions will be opened with
LWR wherever permissible by the provisions of the LWR
manual.
2. All complete track renewals (primary) shall provide for
LWR/CWR wherever permissible. Existing rails on
permitted locations shall be converted into LWR/CWR
provided they meet the requirements laid down in the
Manual for Welding of rail joints by Alumino thermit (SKV)
process.
3. In goods running lines, goods yards, reception yards
marshalling and classification yards, rail joints may be
welded to form LWR if the condition of all components of
the track is generally sound and without any deficiency
subject to such relaxation as may be approved by Chief
Engineer, in each specific case.
R = CURVE RADIUS IN m
P = AXIAL COMPRESSIVE
FORCE IN LWR IN kg
= 2 AEαt
P/R
kg/m
Fig. 4.1 External Equilibrium of Curved LWR Track
R
R
P
P
(47)
4.2. Alignment
1. LWR/CWR shall not be laid on curves sharper than 440
meters radius both for BG and MG.
As indicated in Fig. 4.1 the external equilibrium of a curved
elastic beam of radius R subjected to a longitudinal force ‘P’
requires a continuously distributed external force of magnitude ‘f’
where
m kg
R
P
f / ·
. This will be derived from the lateral ballast
resistance τ and
R
P
− τ is the effective lateral resistance against
buckling danger. In order to ensure that the stability of the LWR in
curve is the same as in straight the lateral ballast resistance in
curve should be made high by at least P/R kg/m. Hence a larger
shoulder width on curves and a restriction on the degree of
curvature is prescribed. On Broad Gauge a shouder ballast width
of 500mm has been adopted.
(48)
Note : Shoulder width of 600mm for 100m on either side of
common tangent point for reverse curves with radius less than
1500m.
Fig. 4.2 : LAYING LWR THROUGH REVERSE CURVES
REVERSE CURVES NOT
SHARPER THAN 2
0
f
100m
600mm
l
600mm
100m
COMMON TANGENT
POINT
f
2. LWR/CWR may be continued through reverse curves not
sharper than 875 metres. For reverse curves sharper than
1500 meters radius, shoulder ballast of 600 mm over a
length of 100 m on either side of the common point should
be provided. These details are shown in Fig. 4.2.
4.3. Gradients :
1. The steepest grade permitted is 1 in 100 for LWR sections.
This is because steeper grades imply larger longitudinal
forces due to traction and braking which would be
detrimental to the health of the LWR causing an increase
in the longitudinal stresses in the rail.
2. A vertical curve shall be provided at the junction of grades
when the algebraic difference between the grades is equal
to or more than 4mm per metre or 0.4% as laid down in
para 419 of IRPWM. This vertical curve will serve to
smoothen the geometrical transition and introduce a
gradual change in the direction of longitudinal force as
shown in Fig.4.3.
3. Vertical curves should be of adequate radius as indicated
in the table below
Broad Gauge Metre Gauge
Route Minimum Radius Route Minimum Radius
A 4000 m AII routes 2500 m
B 3000 m
C, D & E 2500 m
(49)
4.4 TRACK STRUCTURE
4.4.1. Formation :
LWR/CWR should be laid on stable formation. Yielding formations/
troublesome formations should normally be isolated from the LWR
section by laying SEJs on either side. Yielding formations will result
in differential settlement and deformation of track causing
misalignment and unevenness. The stability of the LWR will be
affected and buckling of the LWR may result. The formation width
is critical to avoid rolling down of ballast in banks.
4.4.2. Ballast Cushion and Section :
At least 250mm clean ballast stone cushion should be available
below the sleeper. For speeds above 130kmph on BG and 100
kmph on MG, a 300mm clean ballast cushion or a 200mm clean
ballast cushion with 150 mm of sub-ballast has been prescribed.
The ballast section for the LWR is a heaped up section with heaping
up of 100mm starting from the edge of the sleeper. The shoulder
ballast width is increased to 350mm from the standard shoulder
width of 300mm. In case of curves the shoulder ballast width
prescribed is 500 mm.Details of ballast profile in bank and cutting for
BG & MG in single and double lines are given in Fig.4.4.(a),(b),(c) &
(d).
VERTICAL CURVE
VERTICAL CURVE
l
α1
α2
α1
α2
CHANGE OF GRADE
= α1 + α2
(i) Junction of a rising
& a falling grade
CHANGE OF GRADE
= α1 - α2
(i) Junction of rising or
falling grades
Fig. 4.3 Provision of Vertical curves at grade Intersection Points
(50)
Fig.4.4. (a)
(51)
30 30
6850
6850
6850
6850
6850
6850
6250
6250
6250
6250
6250
6250
Fig.4.4. (b)
(52)
30
30
5850 5250
5850 5250
5850 5250
5850 5250
5850
5850
5250
5250
5850 5250
5850
5850
5250
5250
(53)
Fig.4.4. (c)
30 30
12160 11550
12160 11550
12160 11550
12160 11550
12160 11550
12160 11550
12160 11550
12160 11550
12160 11550
Fig.4.4. (d)
(54)
30 30
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
9210 9810
4.4.3 Sleeper and fastenings :
As explained earlier, the entire theory of the LWR is built on the
premise that there is no relative slip of the rail with respect to the
sleeper, and any movement which occurs is with rails and sleepers
moving together. To achieve this the following types of sleepers
and fastenings have been approved for use in LWR/CWR :-
BG
(i) Concrete sleepers with elastic fastenings.
(ii) Steel trough sleepers with elastic fastenings for speeds
not exceeding 130 kmph (elastic fastenings are used
on steel trough sleepers with modified loose jaws or
with a steel pad plate welded to the steel sleeper.) As
an interim measure speeds upto 160 kmph have been
permitted with such arrangements.
Some exceptions to the above recommendations are as below :
(i) Steel sleepers with 2 way keys and CST-9 sleepers
with 2 way keys are permitted for speeds upto 130
kmph, provided no maintenance problem is faced and
performance is satisfactory.
(ii) On steel trough sleepers with key fastenings, the
breathing lengths shall preferably be provided with
elastic fastenings.
(iii) Special precautions to be taken for CST-9 track are as
under :
1. On single line section keys on adjacent sleepers to
be driven in opposite directions :
2. In double line sections 75% of the keys to be
driven in the direction of traffic and 25% in the
opposite direction. In the breathing length,
however, adjacent sleepers will have keys driven in
the opposite directions.
(55)
Breathing length
Central Portion of double line section
Fig 4.5 Direction of key driving with CST-9 sleepers.
q
q
q
q
Direction of traffic
Breathing length
Central Portion of double line section
Fig 4.6 Direction of key driving with steel sleepers.
Direction of traffic
(56)
The direction of key driving for CST-9 and steel
trough sleepers in LWR territory for single and
double line sections is given in Fig 4.5 and 4.6
respectively.
MG : The recommended sleepers for speeds above 75kmph but a
must for speeds above 100 kmph are :
1. Concrete sleeper with elastic fastenings.
2. Steel trough sleepers with elastic fastenings.
The recommended sleepers for speeds upto 100kmph are:
1. Steel sleepers with two way keys.
2. CST-9 sleepers with keys.
Wooden sleepers with anti-creep bearing plates and two way keys
or elastic fastening may be permitted, for continuing LWR, if
behaving satisfactorily for a maximum speed of 130 kmph in BG
and 100 kmph in MG.
Sleeper density :
The minimum sleeper density (number of sleepers/km) in LWR/
CWR shall be as follows :-
Type of sleeper Sleeper density in BG/MG
i) PRC 1310 in temperature zone I & II
ii) PRC 1540 in temperature zone III & IV
iii) Other sleepers 1540 in all temperature zones
4.4.4 Rails : The following rail sections can be welded to form an
LWR :
BG MG
90R 75R
52kg 90R
60kg
(57)
1. In MG, 60 R rails converted into LWR may be permitted to
continue if showing satisfactory performance.
2. In the same LWR, different rail sections are not permitted.
This is because of the following reasons:
(i) Thermal forces generated in rails of different cross
sectional areas are different. This makes the behavior of
the LWR non uniform. The destressing temperatures are
also different for 52kg and 90R rails.
(ii) While permitting two different rail sections in an LWR,
combination welded joints cannot be avoided. As the
gauge faces have to be matched, eccentricity is induced in
the axial forces, resulting in additional stresses in the rail.
(iii) The ultrasonic flaw detection of combination welds is not
completely foolproof.
The track structure suggested at the junction of a 52kg and
60kg LWRs is shown in Fig.4.7. The suggested track structure
meets with the requirement of continuing the same track
structure for three rail lengths beyond the SEJ.
3. Before converting an existing fish plated track into LWR/
CWR, following precautions should be taken :
Fig.4.7
combination weld
60 kg LWR
60 KG SEJ 52 KG SEJ
60kg 3 Rail panel 52kg 3 rail panel
52kg LWR
j
(58)
i) Rails to be ultrasonically tested and all
defective rails replaced.
ii) Rail ends which are bent, hogged, battered or
having a history of bolt hole cracks should be
cropped before welding. Cropping by 300mm –
450mm is generally done.
iii) Rails should have a residual life of more than
10 years.
4. New rails used in LWR/CWR shall be as far as possible
without fish bolt holes. If the rail ends have to be joined
during installation then use should be made of 1 m long
fishplates/ normal fishplates with screw clamps with SRs of
30 kmph with 1 m long fishplates and 20 kmph with normal
fishplates.
4.4.5 MISCELLANEOUS
1. Level X-ings should not fall within the breathing lengths of
the LWR as level crossings are rigid structures and will not
permit thermal movements of the breathing length to take
place.
2. Normally LWRs are not taken through points and
crossings. For concrete sleeper track a 3-rail panel should
be provided between the SEJ and stock rail joint. Similarly
a 3-rail panel is required to be provided between the heel
of crossing and SEJ. The point and crossing assembly has
a distinct track structure with different maintenance
schedules and subject to large lateral forces when a train
negotiates a turnout. Due to this reason, it was considered
desirable to isolate the point and crossing assembly from
the LWR. However with the introduction of stress frames in
the point and crossing area or taking other special
measures it is now possible to carry an LWR through
station yards including points and crossings.
3. Insulated Joints for track circuiting in LWR/CWR shall be
provided using glued joints of the G3(L) type. The G3 (L)
type glued joint has a pullout capacity of 150 tons for 52
kg rails and 175 tons for 60 kg rails and an insulation
(59)
resistance not less than 25 megaohms in dry condition or
3 kiloohms under saturated conditions. Glued joints are
considered as track components whose procurement has
to be done by the engineering department.
4. Location of SEJs : The exact location of the SEJ will be
fixed taking into account location of various obligatory
points such as level crossings, girder bridges, points and
crossings, gradients and curves. SEJ with straight tongue
and stock rail shall not be located on curves sharper than
0.5 degree (3500m radius). SEJ shall also not be located
on transition portion of cuves.
4.5. LWR on Bridges :
(i) Bridges with ballasted decks (without bearing) :
LWR/CWR can be continued over bridges
without bearings like slabs, box culverts and
arches.
(ii) Bridges with/without ballasted deck with
bearings : When the bridge structure and the
track exhibit movement relative to each other
then there are interaction effects which have to
be taken into consideration. These interaction
effects have been discussed in UIC774-3R
(Reference 9) some parts of which have been
given below.
4.5.1 LWR on bridges: Track Bridge Interaction( UIC 774-3R)
1. Introducing a bridge under a CWR track means effectively
that the CWR track is resting on a surface subject to
deformation and movements hence causing displacement
of the track. Given that both track and bridge are
connected to one another either directly or through the
medium of ballast and are able to move, any force or
displacement that acts on one of them will induce force in
the other.
2. All actions which lead to interaction effects are those that
cause relative displacement between the track and the
deck. These are :-
i) The thermal expansion of the deck only in the
(60)
case of the CWR or the thermal expansion of
the deck and of the rail whenever a rail
expansion device is present.
ii) Horizontal braking and acceleration forces.
iii) Rotation of the deck on its supports as a result
of the deck bending under vertical traffic loads.
iv) Deformation of the concrete structure due to
creep and shrinkage.
v) Effects of temperature gradient.
Out of these 5 factors, the first 3 are more
important.
3. The forces created due to interaction between track and
bridge are dependent on a number of parameters of bridge
and track both :
The bridge parameters affecting the interaction forces are :
(1) Expansion length of the bridge(L): For a single span
simply supported bridge the expansion length is the span
length. For a continuous bridge with a fixed support at the
end, it is the total length of the deck. If the fixed elastic
support is located at some intermediate point, the deck is
considered to have two expansion lengths on either side of
fixed elastic support.
(2) Support stiffness : The resistance of the deck to
horizontal displacement is a fundamental parameter as it
affects all interaction phenomena. This factor is determined
primarily by the total stiffness of the supports. The total
support stiffness is composed of the stiffness of each
support. The stiffness of each support is in turn composed
of the stiffness of the bearing, pier, base, foundation and
soil. The stiffness K of the support including its foundation
to displacement along the longitudinal axis of the bridge is
given by
(61)
( )
( ) cm i
KN H
K
∂ ∑
·
with a h p i ∂ + ∂ + ∂ + ∂ · ∂ φ
where, · ∂p displacement at the head of the support due to
deck’s deformation (this could be calculated assuming the pier to
be a cantilever fixed at the base)
· ∂φ displacement at the head of the support due to
foundation rotation.
· ∂h
displacement due to horizontal movement of the
foundation.
· ∂a
relative displacement between upper and lower parts of
the bearing
The value of the displacement component is determined at the
level of the bearing as shown in Fig 4.8.
3) Bending stiffness of the Deck : As a result of bending
of the deck the upper edge of the deck is displaced in the
horizontal direction. This deformation also generates
interaction forces.
H
h
p
h
f
H
Fig. 4.8
(62)
φ ∂
φ
φ
p ∂ h ∂
4) Height of the Deck : The distance of the upper surface
of the deck slab from the neutral axis of the deck and the
distance of neutral axis from the centre of rotation of piers
affects the interaction phenomena due to bending of the
deck.
TRACK PARAMETERS : The resistance ‘k’ of the track
per unit length to longitudinal displacement ‘u’ is an
important parameter. This parameter in turn depends on a
large number of factors such as whether the track is
loaded or unloaded, ballasted or frozen, standard of
maintenance etc. The resistance to longitudinal displacement
is higher on loaded track than on unloaded track as can be
seen from Fig. 4.9. The value of k has to be established by
each railway system as per its track structure.
Once the values of K, the stiffness of the bridge structure
and k, the stiffness of the track have been evaluated, use
can made of the interaction diagrams given in UIC774-3R
for calculation of the additional stresses in the rail and
additional forces at the bridge support due to each of the
actions causing interaction effects: namely (1)change of
temperature (2) acceleration and braking forces (3) deck
deformation.
u
Stiffness
‘k’
Resistance of rail on sleeper
(unloaded condition)
k = 40 KN/m
Resistance of rail on sleeper
(loaded condition)
k = 60 KN/m
u = 0.5mm
Fig. 4.9 TRACK STIFFNESS PARAMETERS
(FROZEN BALLAST)
(63)
1 Changes in temperature :
It is assumed that there is change of temperature of
C
0
35 t
from
the reference temperature for the bridge while for the rail could
deviate by
C
0
50 t
. Due to change of temperature additional
stress will develop in the rail and additional force at the support.
These are obtained from the interaction charts given in UIC774-3R.
2. Actions due to braking and accleration
The braking and accleration forces applied at the top of the rail are
assumed to be distributed over the length under consideration
with the following standard values :
Accleration = 33 KN/m per track
Braking = 20 KN/m per track
These values could be modified to take into account the
longitudinal loadings given in the Bridge Rules.
3. Actions due to bending of deck
Vertical traffic loads cause the deck to bend, which in turn causes
rotation of the end sections and displacements of the upper edge
of the deck. The design curves for the evaluation of the interaction
due to vertical bending of the bridge deck have been evaluated
u
S
t
i
f
f
n
e
s
s


k
’ Resistance of rail on sleeper
(unloaded condition)
k = 12-20 KN/m
Resistance of rail on sleeper
(loaded condition)
k = 40 KN/m
u = 2mm
Fig. 4.9 TRACK STIFFNESS PARAMETERS
(NORMAL BALLAST)
(64)
with respect to the standard longitudinal plastic shear resistance
equal to 20 KN/m and 60 KN/m for unloaded and loaded track
respectively.
The design curves are given for the following two different
situations :
- deck bridge – the track lies on the top of the bridge
deck (deck neutral axis below track
axis)
- through girder bridge – deck neutral axis above track axis.
Combining load cases
For calculation of the total support reaction and in order to
compare the global stress in the rail with the permissible value set
by each railway, the global effect

R is calculated as follows :
( ) ) ( ) ( bending R braking R T R R γ β α + + ∆ · ∑
γ β α , , are the combination factors.
Permissible additional stresses in continuous welded rail on the
bridge
Theoretical stability calculations on UIC 60kg CWR of a steel grade
giving at least 900 N/mm
2
strength, minimum curve radius 1500 m,
laid on ballasted track, with concrete sleepers and consolidated
ballast cushion greater than 30 cm give a total possible value for
the increase of rail stresses due to track/bridge interaction as
indicated below:
The maximum permissible additional compressive rail stress is 72
N/mm
2
,
The maximum permissible additional tensile rail stress is 92 N/
mm
2
.
For structures consisting of one deck, the values of the interaction
effects can be calculated by using the design graphs in Appendix
‘A’ – page 36 and Appendix ‘B’ page 42 give in UIC report 774 –
3R.
(65)
4.5.2. Provisions given in the LWR Manual for carrying
LWR over bridges:
However, for a simple understanding of the problem let us
consider the effect of thermal variation alone as the cause of
interaction between the girder and the LWR. As a result of thermal
variation the girder has a tendency to expand or contract being
provided with bearings. On the other hand the central portion of
the LWR is fixed in position irrespective of the temperature
changes that occur. This results in an interplay of forces between
the girder and the LWR, the magnitude of the force being
dependent upon the nature of fastenings being provided between
the rail and sleeper. To clarify this aspect of interplay of forces
between rail and girder, consider the case of a girder bridge
provided with fastenings between the rail and sleeper with a creep
resistance equal to ‘p’ kg per rail seat. The bridge sleepers are
rigidly fixed to the top flange of the girder by means of hook bolts.
On variation of temperature due to the creep resistance of the
fastenings, free expansion/contraction of the girder is prevented.
Consequently additional forces are developed both in the girder as
well as in the rail. The magnitude of this force developed depends
upon the value of ‘p’( the creep resistance) and orientation/nature
of the bearings provided in each span of the bridge.
The following cases have been considered:
Single span bridge : 1. One end fixed, other end free.
2. Both ends of girder with free bearings.
Multiple span bridge: 1. One end fixed and the other free with
dissimilar bearings on a pier
2. One end fixed and the other free with
similar bearings on a pier
3. free bearings at both ends.
The forces developed in the rail and girder for each of the five
cases mentioned above are given in Fig. 4.10. These LWR force
diagrams indicate that :
i) For sliding bearings at both ends of the girder, the increment of
force in the LWR is
4
np
, where ‘n’ is the number of sleepers per
span with creep resistant fastenings and ‘p’ is the creep resistance
(66)
Fig. 4.10 (a)
n = No. of sleepers per span
p = creep resistance per rail seat
Fig. 4.10 (b)
n = No. of sleepers per span
p = creep resistance per rail seat
(67)
← ← ← → → →
→ → → → → →
rise
rise
Fig. 4.10 (d)
Fig. 4.10 (c)
m = No. of spans
n = No. of sleepers per span
p = creep resistance per rail seat
n = No. of sleepers per span
p = creep resistance per rail seat
(68)
→ → → → → → → → → →
← ← → →← ←→ → ← ←→ →
p
np
per rail seat (4.10(a)). This increment of force will remain the same
irrespective of the number of spans of the bridge (4.10(d)).
ii) In girders with one end fixed and the other end free the
increment of force in the LWR at the roller end is
2
np
for a single
span bridge, where n = number of sleepers in the span with creep
resistance of ‘p’ kg per rail seat (4.10(b)). If it is a multiple span
bridge with ‘m’ number of spans, the increment of force in the
LWR at the roller end will be
2
p n m × ×
. The resultant LWR force
diagram is shown in the sketch (4.10(c)). This is the case when on
a pier bearing for one girder is a fixed bearing while the bearing of
the other girder is a free bearing.
iii) There could be a situation where a pier supports similar
nature bearings i.e. the bearings of the two girders are either fixed
or free. In this case there will be no cumulative build up of force
and the resultant LWR force diagram will be as indicated in
Fig.4.10(e).
Fig. 4.10 (e)
n = No. of sleepers per span
p = creep resistance per rail seat
(69)
← ← ← → → → ← ← ←

→ ←
np
In order to avoid interplay of forces between the LWR and
girder a possible solution would be to provide rail free
fastenings between rail and sleeper on the girder bridge. It is
with this assumption that the provisions for laying an LWR over
bridges have been framed in the LWR manual.
Fastenings used to connect the rail to the sleeper could be of
two types : (1) Creep resistant fastenings and (2) Rail free
fastenings which are now termed as zero longitudinal restraint
fastenings. RDSO Report No. C-169 investigates the creep
resistance offered by different types of rail –sleeper fastenings.
On the Indian Railways we have been traditionally using dog
spikes and rail screws as rail free fastenings although now
Pandrol has come up with a zero longitudinal restraint
design(Fig 4.11). Under normal circumstances there is a small
gap between the base plate (steel) and the top side of the rail
foot. In case of large lateral forces, the baseplate prevents the
overturning of the rail. The pad under the rail is made up of low
friction material like teflon, which provides an almost zero friction
Fig. 4.11
(70)
movement between the rail and sleeper.
Use of rail free fastenings on bridges where LWR is proposed
to be used is now mandatory due to requirement of minimizing
the interaction of forces between the LWR and the girder.
However, this results in another problem : enhanced gap at
fracture, when the fracture occurs on the approach of bridge
laid with LWR.
Consider an LWR laid on normal formation with the usual force
diagram A B C D. in the event of fracture at location ‘F’ the
stress in the LWR is released at that location and two new
breathing lengths B
1
F and C
1
F are formed on either side of the
fracture location. (Fig 4.12)
Fig. 4.12
B
A
F D
C
C’ B’
F
(71)
Fig. 4.13
A E D
F
H
G
F
B
L0
C
The gap g
1
at the fracture location will be given by
( )
2
2
|
2
1
× ·
R
t AE
g
α
———— (1)
[Assuming equal movement on either side of F]
R
|
represents the longitudinal ballast resistance mobilised at
the time of the fracture, which is generally about 50% to 60%
of the normal R value, due to the sudden nature of occurrence
of a fracture.
However, if the same fracture had occurred in the approach of
a bridge provided with LWR and rail free fastenings the
modification of the force diagram will be as given in the figure
4.13.
In this figure, ABCDEFGH represents the altered force
diagram.
Gap at fracture in this case will be
( )
t L
R
t xAE
g . .
2
2
0 |
2
2
α
α
+ · ————— (2)
Where L
0
is the span length of the bridge provided with rail
free fastenings.
Expressions (1) and (2) indicate that the gap at fracture is
enhanced by an amount equal to t L α
0
, when a girder bridge with
rail free fastenings is located in the central portion of the LWR.
Indian Railways have fixed the permissible gap at fracture as
50mm where by expression (2) becomes
( )
mm t L
R
t AE
x 50
2
2
0 |
2
< + α
α
This expression is applicable for both BG and MG tracks.
However, as the wheel diameter of MG stock is smaller than
(72)
BG, the fracture gap of 50 mm is more critical for MG.
Over the years attempts have been made to increase the
value of L0 by adopting various techniques :-
(1) One way could be to increase the value of R, the
longitudinal ballast resistance mobilized at the fracture. This
could be done by :-
·l Compacting the ballast in shoulders and cribs of the bridge
approach sleepers.
·l Enhancing the sleeper density to 1660 Nos./km in the bridge
approach.
·l Heaping up of ballast in the bridge approach starting from the
foot of the rail.
·l Box anchoring sleepers wherever required.
These measures have to be taken in the bridge approaches
50m on either side.
Table 1 of the LWR Manual 1996 gives the maximum overall length of
girder permitted on LWR/CWR in with the following stipulations :
1. Girder bridge should have sliding bearings on each end with
single span limited to 30.5m.
2. Rail should be provided with rail free fastenings throughout
the length of the bridge from abutment to abutment.
3. The approach track should be suitably upgraded as mentioned
above.
2) Another way of increasing the value of Lo would be to improve the
approaches as mentioned above in addition to providing a few sleepers
on each span with creep resistant fastenings. The creep resistant
fastening will hold the rail and prevent the gap at fracture from becoming
excessive.
However, provision of creep resistant anchors implies an interplay of
forces between the rail and grider. Hence the following stipulations are
made for bridge provided with rail free fastenings and partly box-
anchored (with single span not exceeding 30.5m and having sliding
bearings at both ends).
(73)
(1) On each span 4 central sleepers will be provided with creep
resistant fastenings and remaining sleepers with rail free
fastenings.
(2) Bridge timbers laid on girders shall not be provided with through
notch but shall be notched to accommodate the individual rivet
heads.
(3) The girders shall be centralized with reference to the location
strips on the bearing before laying LWR/CWR.
(4) The sliding bearings shall be inspected twice a year and oiling
and greasing of the bearing carried out once in two years.
Table 1 of LWR Manual
Temperature Rail Rail free fastening Rail free fastening
Zone Section on bridge on bridge and partly
box-anchored/
creep resistant
fastenings
Type of sleeper Type of sleeper
used in approaches used in approaches
PRC/ST PRC/ST
I 60kg 30m 77m
52kg/90R 45m 90m
II 60kg 11m 42m
52kg/90R 27m 58m
III 60kg 11m 28m
52kg/90R 27m 43m
IV 60kg 11m 23m
52kg/90R 27m 43m
The LWR Manual has also suggested some additional methods of
carrying an LWR over bridges. These are discussed below :
(1) Providing an SEJ on each pier with rail free fastenings on the
bridge. In order to avoid creep four sleepers on each span will
be box-anchored. These sleepers will be at the fixed end of
the girder, if the girder is having rollers at one and rockers on
the other side. These sleepers will be at the centre of the span
if the girders are having sliding bearings on both sides. This
arrangement is shown in Fig.4.14.
(74)
(75)
(2) Providing an SEJ at the far end approach of the bridge using
rail free fastenings over the girder bridge(Fig 4.15):
In this arrangement an SEJ is provided at the far end approach
of the bridge(abutment away from the LWR) at a distance of
10m away from the abutment with rail free fastenings on the
bridge proper.The SEJ will have to cater to the free expansion
or contraction of the rail on the bridge as well as movement of
the breathing length.Hence the SEJ will have to be a wide gap
SEJ capable of accomodating larger movements.The
permissible span lengths with normal SEJs and 190mm
maximum gap SEJs are given on the adjoining page.
M
a
x
.

l
e
n
g
t
h

o
f
I
n
i
t
i
a
l

g
a
p

t
o

b
e
b
r
i
d
g
e

w
i
t
h

S
E
J
p
r
o
v
i
d
e
d

a
t

t
d
R
a
i
l
M
a
x
.
m
o
v
e
w
i
t
h

S
T
/
w
i
t
h
w
i
t
h

S
T
/
w
i
t
h

C
S
T
-
9
t
e
m
p
.
m
e
n
t

o
f
P
R
C
C
S
T
-
9
P
R
C
a
p
p
r
o
a
c
h
Z
o
n
e
S
E
J

(
m
m
)
a
p
p
r
o
a
c
h
a
p
p
r
o
a
c
h
a
p
p
r
o
a
c
h
s
l
e
e
p
e
r
s
s
l
e
e
p
e
r
s
s
l
e
e
p
e
r
s
s
l
e
e
p
e
r
s
I
V
1
9
0
5
5
m
4
5
m
7
.
0

c
m
6
.
5

c
m
I
I
I
1
9
0
7
0
m
7
0
m
7
.
0

c
m
6
.
5

c
m
I
I
1
9
0
1
1
0
m
1
0
0
m
6
.
5

c
m
6
.
5

c
m
I
1
9
0
1
6
0
m
1
5
0
m
6
.
5

c
m
6
.
0

c
m
I
I
1
2
0
2
0
m
1
5
m
4
.
0

c
m
4
.
0

c
m
I
1
2
0
5
0
m
5
0
m
4
.
0

c
m
4
.
0

c
m
N
o
t
e

:

S
E
J

i
s

t
o

b
e

i
n
s
t
a
l
l
e
d

1
0

m
e
t
r
e

a
w
a
y

f
r
o
m

t
h
e

a
b
u
t
m
e
n
t
s
.
(76)
(77)
F
i
g
.

4
.
1
4
F
i
g
.

4
.
1
5
CHAPTER V
LAYING AND MAINTENANCE
5.1 Laying of LWR
1. An initial survey of the section where the LWR/CWR is
proposed to be laid should be carried out. A foot by foot
survey is recommended.
2. Locations where LWR/CWR cannot be laid, need to be
identified. These locations could be:
i) Sharp curves
ii) Bridge locations
iii) Steep Gradients
iv) Points and crossings
v) Troublesome formations
vi) Distressed bridges
These locations will be isolated from the LWR by providing
SEJs on either side.
3. A detailed plan shall be made showing the exact location
of the SEJs and various other features such as level
crossings, yards, curves, points and crossings, gradients
and bridges. This is called the LWR plan. This plan should
be got approved by the T.H.O.D. if there are no deviations
from the provisions of the LWR manual or by CE/CTE if
any deviations are proposed.
4. Temperature Records : The LWR Manual (1996)
prescribes that each railway should nominate 8 to 10
stations on its jurisdiction where temperature records over
a period of 5 years should be built up by installing suitable
continuous recording thermometers. The maximum and
minimum rail temperature for a continuous period of at
least 5 years shall be ascertained and the mean rail
temperature for the region arrived at.
This could provide the basis for fixing the rail neutral
temperature, ascertaining the periods during the year when
(78)
maintenance operations could be carried out or hot
weather and cold weather patrolling need to be introduced.
If these records are not available, use may be made of the
rail temperature map given in the LWR Manual.
5. Materials required for laying an LWR :
i) Four numbers of 6.5 metres or longer rail
pieces of the same rail section as the LWR.
ii) 2 sets of SEJs with sleepers and fastenings.
iii) Adequate number of 1 meter long fishplates
with screw clamps/joggled fish plates with
slotted grooves and bolted clamps.
iv) Rail closures.
v) Rail cutting equipment.
vi) Welding equipment.
vii) Destressing equipment i.e. roller, tensor,
wooden mallets.
6. Preliminary works to be carried out prior to laying the LWR :
i) Replacement of insulated joints by glued joints if
LWRs are to be laid in station yards.
ii) Realignment of curves.
iii) Lifting or lowering of track to eliminate sags and
humps.
iv) Introduction and improvement of vertical curves.
v) Stabilization of troublesome formation.
vi) Rehabilitation of weak bridges.
7. Welding of rails to form LWR :
i) Rails are supplied from workshop flash butt
welding plants in 3-rail, 10-rail, 10-rail or 20-rail
panels. Care should be taken while handling
90 UTS rails at the time of unloading.
ii) The 10-rail/15-rail/20-rail panels should be
placed in the track and subsequently welded to
form the LWR of required length.
(79)
iii) Two complete sets of SEJs shall be inserted at
predetermined locations with gaps in the mean
position. Closure rails of 6.5 metres or longer
length shall be provided on either side of SEJs
to facilitate adjustment of gaps during
destressing operation.
iv) Laying of welded panels or welding of joints at
site can be done at any time of the year. But
after welding sufficiently long panels of about 1
km length or longer, destressing of the LWR
should be undertaken as soon as possible.
v) Speed restrictions (30 kmph) should be
imposed if fishplated joints are existing in an
LWR. If a temporary joint in the form of a 1m.
long fishplate with screw clamps or a joggled
fishplate/normal fishplate with screw clamps
exists, then speed restrictions of 30 kmph and
20 kmph respectively should be imposed. A
watchman should also be posted at a clamped
joint.
5.2 Destressing Operations
5.2.1 One of the most important maintenance operations in the
LWR is destressing the LWR. Destressing is the operation of
removing the locked up stresses in the LWR and bringing the LWR
to a stress-free state at a predecided temperature called the
stressfree temperature. It is also called the neutral temperature. As
per the LWR Manual, the stress-free temperature should lie
between t
m
+ 5
0
C and t
m
+ 10
0
C for 60kg and 52kg rail sections
and between t
m
and t
m
+ 5
0
C for 90 R rail sections where tm
denotes the mean rail temperature.
5.2.2 Periodicity of destressing and conditions which would
warrant destressing to be done :
Following the Khanna accident, a periodicity of once in 3 years for
destressing every LWR had been prescribed. Subsequently these
instructions were withdrawn and current instructions do not lay
down any periodicity for destressing. The LWR Manual lays down
that destressing should be done :
(80)
1. (i) when gap at SEJ goes beyond the prescribed limits.
(ii) when tongue rail/ stock rail cross the mean position.
2. After a special maintenance operation like deep screening.
3. After restoration of track following an unusual occurence.
4. If the number of locations where temporary repairs have
been done exceed 3 per kilometer.
It is suggested that the condition of the entire LWR should
be considered before taking a decision of destressing.
5.2.3 Destressing operations could be done in two ways :
(1) Manually
(2) Using a tensor
1. Manual Destressing of the LWR : When the prevailing
temperature tp falls in the range prescribed for the stress-free
temperature then manual destressing of the LWR can be resorted
to. Work should be done in the presence of a PWI.
The steps involved are given below :
A) Pre- Block Activities:
1) All impediments to free rail movement such as
check rails, rail anchors etc. should be removed.
2) 4 Nos. of closure rails to be created at either
end of the LWR, next to the SEJs.
3) ERC clips should be greased so that their
removal during the block will be easy.
4) Prior to the block, 50% of the ERCs are
loosened and a speed restriction of 30 kmph
imposed.
B) Block Activities:
1) Arrange for block of adequate duration taking
into account the length of the LWR to be
destressed and the labour available.
(81)
2) The closure rails are removed and placed on
the cess at either end of the LWR.
3) ERCs are removed starting from the ends
towards the centre. The rails are subsequently
lifted off the rail seat and supported on rollers at
every fifteenth sleeper. Side rollers should be
used for curves.
4) The rails are tapped with a wooden mallet to
remove any builtup stress and subsequently
lowered on the rail seat after removing the
rollers. This is a good opportunity to change the
rubber pads.
5) ERCs are put back starting from the middle
towards the ends. During this operation the
temperature of the rail should lie within the
prescribed range for the destressing or stress-
free temperature.
6) The SEJs at either end are adjusted for a
standard 40 mm gap at the destressing
temperature. Suitable cuts in the closure rails
can be made and inserted back in the track
after which the block can be removed.
7) Welding of the closure rails can be performed
if required in a separate block. Fig 5.1 shows
schematically the operations involved in
destressing manually with the options available.
2. Destressing using tensor :
When the prescribed temperature range for the neutral
temperature is not available in the prevailing conditions, the rail
tensor can be used for destressing operations. The rail tensor
is a hydraulic or mechanical device which can create tensile
stress in the rail by pulling the same.
Principle : The rail tensor creates tensile stress in the rail of such
magnitude that these stresses completely balance the
compressive stress created in the LWR when the temperature
rises from t
P
(the prevailing rail temperature) to t
n
, the defined rail
(82)
neutral temperature. The LWR is made stress-free at t
p
by
removing the fastenings and subsequently tensile stresses are
created in it by pulling the LWR by a calculated amount :
Compressive stress created in the LWR due to rise of temp
from t
p
to t
n
with zero stress at t
P
= Eα(t
n
– t
p
)
The tensor creates the same amount of tensile stress by pulling
the LWR section by an amount equal to ∆.
If the length of the section is L, then strain
L

·
The tensile stress due to this strain will be
L
E

) (
p n
t t E
L
E − ·

∴ α
or ) (
p n
t t L − · ∆ α
This expression gives the extension to be imparted to a
segment of length ‘L” of the LWR to get the stress-free
temperature at t
n
.
Equipment : The tensor is basically a pulling device which
could be hydraulic or mechanical. It is non-infringing, enabling
trains to pass when in position. It grips the web of the rail
using a special cam arrangement eliminating bending, dipped
joints and rail-head marks. It can be dismantled into different
parts for being transported, the weight of the heaviest
component being 54 kg. (Details of a tensor are given in Fig
5.2) Some of the details of a tensor manufactured by CTR
Industries, Pune are given below:
Total weight = 325 kg (without pump)
Pulling force = 70 tonnes
Pushing force = 30 tonnes
Stroke = 380 mm
Overall dimensions : 4500 mm length
1200 mm width
(83)
Note : 1. Rail cuts at locations 1, 2,3,4,5,6,7,8 to create two
pairs of closure rails.
2. Fastenings removed from ends towards centre.
3. Fastenings put-back starting from centre towards ends.
Fig. 5.1(a) Destressing Manually
Note : 1. LWR has been split into three convenient sections.
2. Cuts made as indicated prior to block.
3. Level crossing used as a convenient location to split
LWR.
Fig. 5.1(b) Destressing Manually
By Splitting Long LWRs.
Level crossing
(84)
Steps in destressing by the rail tensor (Fig 5.3) :
1) The destressing operations have to be carried out
when the prevailing rail temperature t
p
is less than
the the designed neutral temperature.
2) Make cuts 1m apart at the centre of the LWR.
3) Erect marker pillars W0, W1, W2 at convenient
distances of say 100m. The first marker pillar W0
will be erected at a distance equal to the anchor
length from the SEJ. The anchor length is the
length of track where the fastenings will not be
removed during the destressing operations. This
length is required to resist the pull applied by the
tensor. An estimate of the anchor length is as
under:
For BG ..2.5 m per
o
C of t
n
-t
p
For MG ..4.5 m per
o
C of t
n
-t
p
This anchor length is excluded from the
destressing operations and will have to be done
manually. The last marker pillar will be erected at
the point where the LWR is being cut to make a
gap of 1m. These marker pillars may be made on
both the sides of the track. Transfer W
0
on to the
foot of the rail.
4) Unfasten the fastenings starting from the end or
from the centre where the rail is cut. Place the rail
on rollers. Apply pull by tensor to get some
movement at W0 and release. Note movement at
W
0
. If the movement is away from tensor, the zero
correction is positive. Let it be Y
0
5) Transfer the remaining points W1, W2, ....... Wa on
to foot of rail and note down the prevailing rail
temperature tp. Calculate the extentions to be
given in each segment. Extention to be given to
first segment WoW1 will be,
= Y
0
+ (W
0
W
1
)x α x(t
n
- t
p
)
(85)
PART LIST
Sr. ITEM QTY. Sr. ITEM QTY.
No. Nos. No. No.
1. YOKE 2 7. CLEVIS 2
2. LEVER ARM 4 8. PIN WITH CHAIN 6
3. HYDRAULIC 2 9. HOSE 4
CYLINDER PIPE-2m
4. JAW 4 10. HOSE PIPE-4m 2
5. TIE BAR 2 11. HYDRAULIC 1
PUMP UNIT
6. FORK END 4 12. T CONNECTOR 2
Fig. 5.2 HYDRAULIC RAIL TENSOR
11
10
12
9
6
7
5
8
2
1
3
4
(86)
Fig. 5.3
(87)
t
n
This distance is marked on the foot of the rail at
W1 away from the tensor.
Similarly, extention to be given to segment W1W2 =
Extention to segment W0W1 + W1W2 x α x (t
n
-t
p
)
This distance is marked at W2 away from the
tensor.
6) Apply pull by tensor to get the required extention of
segment WoW1.This can ensured by bringing the
mark of required extension opposite to the mark on
the marker pillar W1. Once the required extension
has been given to a segment the segment can be
lowered on to the rail seat and fastenings put back.
7) This process could be continued on to the other
side of the tensor as well either simultaneously or
after tackling one side of the LWR.
8) At the end in order to replace the 1 metre long
closure rail at the centre make a paint mark at a
distance of (6.5m + 2 welds) measured from one
rail end across the tensor. Remove the tensor and
normalize the block.
9) In another block make a cut at the earlier made
paint mark and put in a closure rail piece of 6.5m.
One gap can be welded immediately. The other gap
could be welded also, if 25mm, otherwise use the
tensor to make the gap 25mm and do the welding.
5.3 Regular maintenance operations
5.3.1. Regular track maintenance operations as defined by
the LWR Manual are :
1) Tamping/Packing
2) Lifting
3) Aligning including minor curve realignment
4) Shallow screening/ shoulder screening.
5) Renewal of fastenings requiring lifting of rail.
6) Maintenance of SEJ/buffer rails.
(88)
5.3.2. Most of the operations involve disturbance of the
ballast bed leading to loss of the sleeper to ballast
resistance. Consolidation of track is the process of
building up of sleeper to ballast resistance either
initially before laying of LWR or making up subsequent
loss of resistance by any one of the following (Para
1.18 of LWR Manual)
I (i) For track structures consisting of sleepers other than
concrete sleeper :
(a) Passage of al least 3,00,000 gross tons of traffic on
BG or at least 1,00,000 gross tons of traffic on MG
when compaction of ballast is done using hand-
operated compactors/ consolidators or rammers.
(b) Passage of at least 50,000 gross tons of traffic on BG
or at least 20,000 gross tons of traffic on MG or a
period of 2 days whichever is later, when compaction
is done by mechanized shoulder and crib compactor.
(ii) For track structure consisting of concrete sleepers :
Passage of 50,000 gross tons of traffic on BG or
20,000 gross tons of traffic on MG or a period of 2
days whichever is later.
II One round of stabilization by Dynamic Track Stabiliser (DTS).
III For newly laid LWR/CWR at least three rounds of packing, last
two of which should be with on track tamping machines.
5.3.4 Temperature Restrictions while carrying out
Maintenance Operations in LWR territory :
(i) Normal maintenance operations should be
carried out well before on set of summer. The
temperature range within which such operations
should be performed should be restricted to
t
d
+10
0
C to t
d
–30
0
C where t
d
is destressing
temperature.
(ii) If the rail temperature rises above t
d
+ 20
0
C
during the period of consolidations which has
been defined earlier (para 1.18 of LWR Manual)
then the following steps have to be taken :
(89)
(A) For other than concrete sleeper track :
1. A speed restriction of 30 kmph in BG and
20 kmph in MG will have to be imposed if
mechanical compaction of ballast has not
been done during the maintenance
operations in addition to posting of a
mobile watchman.
2. A speed restriction of 50 kmph in BG and
40 kmph in MG will have to be imposed if
shoulder and crib compaction has been
done during the maintenance operations.
(B) For concrete sleeper track :
1. A speed restriction of 50 kmph in BG and
40 kmph in MG will have to be imposed.
5.3.5. Manual Through-packing of sleepers in LWR
territory :
It must be remembered that till off track tampers are made
available manual packing of concrete sleeper track with crowbars
has been permitted (para 1408 of IRPWM). However, due to the
necessity of opening out of ballast, for undertaking manual
packing, the following precautions have to be observed:
i) Only 30 sleepers should be opened out at a
time, leaving the next 30 sleepers fully boxed
and packed. The intervening sleepers can be
opened out for packing after 24 hours if the GMT
on the BG section is more than 10 and after 48
hours for BG sections with GMT less than 10 or
in MG sections.
ii) Only in case of emergencies and in the presence
of PWI can 100 sleeper spaces be
simultaneously opened out observing the usual
temperature restrictions.
5.3.6. Other instructions while carrying out track maintenance in
LWR territory:
(i) Special attention shall be paid to
(90)
maintenance of track at the following locations :
- SEJs/breathing lengths.
- Approaches of level crossings, points &
crossings and unballasted deck
bridges.
- Horizontal and vertical curves.
(ii) Special attention shall be paid to
maintenance of fastenings in LWR/CWR
especially on concrete sleepers .
(iii) Ballast section shall be properly maintained
paying special attention to locations where
the ballast profile could get disturbed such
as at level crossing approaches, bridge
approaches, curves, pedestrian and cattle
crossings. In order to prevent rolling of
ballast down the slope, the cess should be
properly maintained with regular cess
repairs. Dwarf walls should be provided
wherever track trespass is anticipated to
prevent loss of ballast.
(iv) Adequate ballast should be arranged
before going in for any maintenance
operation such as lifting. Ballast should be
recouped well before the onset of summer.
Ballast procurement being a long lead item,
adequate planning should be made for
timely procurement of this vital item.
(v) While slewing of track using crowbars, care
should be taken to avoid simultaneous
lifting of track.
5.3.7 Mechanised Track Maintenance :
(i) Maintenance tamping using track machines
can be done in continuation from one end of
the section to the other. General lift should
not exceed 50 mm for concrete sleepers
and 25 mm for other than concrete
sleepers.
(91)
Rail temperature restrictions are the same as laid
down for manual track maintenance.
(ii) Lifting of track where needed, in excess of
50 mm for concrete sleepers and 25 mm
for other sleepers shall be carried out in
stages with adequate time gap in between
the successive stages, such that full
consolidation of the previous stage is
achieved prior to taking up the subsequent
lift.
5.3.8 Casual renewal of sleepers, rails and fastenings :
The following precautions should be observed :
(i) Casual renewal of sleepers : Not more than
one sleeper in 30 sleepers shall be
replaced at a time. Should it be necessary
to renew two or more consecutive sleepers
in the same length, they may be renewed
one at a time after packing the sleeper
renewed earlier duly observing the
temperature limits.
(ii) Casual renewal of fastenings :
a. When fastening renewal does not
require the rail to be lifted, fastenings
of not more than one sleeper at a time
shall be renewed at a time, while at
least 15 sleepers in between shall be
kept intact. Work shall be done under
the supervision of a keyman.
b. Fastening renewal requiring lifting of
the rail such as replacement of the
grooved rubber pads shall be done in
the presence of the gang mate with at
least 30 sleepers in between to be
kept intact.
5.3.9. Maintenance of SEJs and buffer rails :
i) SEJs should be checked, packed and aligned
once in a fortnight. Oiling and greasing of the
(92)
tongue rail and stock rail should be done
simultaneously.
ii) Buffer rails are free rails placed in lieu of
Switch Expansion Joints.
(a) In rail temperature Zone I and II, 3 Buffer
rail pairs, while in zone III and IV, 4
buffer rail pairs shall be provided. On
BG, the buffer rail shall be 6.5 m long
while for MG it will be 6.0 meters.
(b) Buffer rail joints are lubricated twice a
year when the rail temperature is
between t
d
+ 15
0
C and t
d
– 15
0
C. A
standard gap of 7.5 mm is provided at a
buffer rail joint at temperature t
d
.
(c) In zone III and IV, if the gap closes at a
temperature lower than t
d
+ 30
0
C or
opens out to 15 mm at a temperature
higher than t
d
– 30
0
C it would indicate :
i) Defective initial gaps.
ii) Inadequate packing in breathing
length.
iii) Creep of LWR.
iv) Movement of rail over sleeper in
breathing length.
In zone I & II the lower
temperature limit for gap fully
opening out and the upper
temperature limit for gap fully
closiing shall be tacken as td-
25
0
Cand td+25
o
C respectively.
In such cases, the LWR should be destressed and the gap at the
buffer rail joints restored to 7.5 mm at td.
5.4 Special Track Maintenance :
These works are generally carried out with a speed restriction in
force.
This would include the following items :
(93)
i) Through fitting renewal
ii) Deep screening/ Mechanised cleaning of
ballast.
iii) Major lowering/ lifting of track
iv) Major realignment of curves.
v) Sleeper renewal other than casual
renewal.
vi) Formation rehabilitation.
5.5. Deep screening in LWR territory :
Provisions laid down in Para 238 of IRPWM will also apply
mutatis mutandis to LWR/CWR. This implies that
1. Work will be done with an SR of 20 kmph in the
presence of a PWI.
2. While tackling two sleepers simultaneously, there
should be at least 4 intermediate sleepers fully packed
and boxed.
3. Deep screening could be carried out in continuation
from one end of the section to the other in the above
manner.
The temperature restrictions are as under :
1. Work should normally be done in the rail temperature
range of t
d
– 20
0
C to t
d
+ 10
0
C.
2. If there is a possibility of rail temperature rising above
t
d
+ 10
0
C during the execution of work then temporary
destressing at a temperature 10
0
C below the
anticipated maximum temperature should be carried
out. This will keep the track in the safe zone as far as
development of high compressive forces is concerned.
If the temperature were to fall by more than 30
0
C
below the temporary destressing temperature, cold
weather patrolling should be introduced.
3. Temporary destressing should be done again after 15
days if there is wide fluctuation of temperature and
there is a possibility of temperature rising further
beyond the anticipated maximum temperature.
(94)
(95)
4. Once the deep screening is completed, then the entire
LWR should be destressed to bring the destressing
temperature to the normal range.
CHAPTER VI
UNUSUAL OCCURENCES IN LWR, INSPECTION &
RECORD KEEPING
6.1 Introduction: This chapter is devoted to a discussion on
various unusual occurrences which may occur in an LWR such as
fractures and buckling, and remedial measures to be taken in the
event of the same.
6.2 Fractures
6.2.1 Rail and weld fractures occur with increasing frequency on
LWR sections in winter due to development of longitudinal tensile
forces as rail temperatures fall below the destressing temperature.
The increasing incidence of rail fractures in LWR sections could
also be attributed to the fact that the rail distressing temperature
has been fixed between t
m
+ 5
0
C to t
m
+ 10
0
C for 52 kg and 60 kg
rails increasing the tensile force created in the rail as temperature
drops towards t
min
.
Causes of fracture : Apart from excessive tensile thermal forces
which could arise in a rail in LWR sections, fractures could occur in
a rail due to variety of causes :
1. Material defects originating during the manufacturing
process such as clusters of non–metallic inclusions,
hydrogen flakes, rolling marks, guide marks etc. which may
be present in spite of the non–destructive tests carried out
on the rails during their quality assurance examination.
2. Residual stresses induced during manufacture : cooling,
rolling, straightening etc.
3. Defects occuring due to incorrect handling of rails e.g.
plastic deformation, scoring, denting etc.
4. Defects associated with faulty welding.
5. Dynamic stresses caused by vertical and lateral loads
particularly by vehicles with wheel flats or when the vehicle
runs over poorly maintained rail joints etc.
(96)
6. Fractures due to corrosion at rail seat and liner location etc.
7. As discussed earlier excessive tensile forces in the rail
generated due to temperature changes of the LWR.
6.2.2 Repairs to Fractures
1. Once a fracture has occurred, the railway official detecting
the fracture should take immediate steps to block the
section and prevent any train movement over the fractured
portion.
2. This done, fracture repairs are done in the following stages :
A. Emergency Repairs
These repairs are carried out to pass the train over the fractured
rail. The following steps are involved :
1. If the gap at the fracture is less than 30 mm, a fishplate for a
rail fracture or a joggled fishplate for a weld fracture is fixed,
using 4 tight screw clamps without a rail closure piece.
2. If necessary, a wooden block may be inserted below the rail
to support the fractured joint.
3. If the gap at fracture is more than 30 mm, a rail–closure
piece will be inserted into the gap after which fishplates or
joggled fishplates will be fixed.
4. This done, the train is allowed to move over the joint with a
SR of stop dead and 10 kmph for the first train and 20 kmph
for the subsequent trains. The LWR Manual authorises a
keyman / gangman to pass the train in such an emergency.
If one meter long fish plates are used during the repair the
SR will be 30 kmph.
B. Temporary Repairs
This essentially involves the removal of the fractured rail from
the track and replacing it with a sound rail–closure piece of
length generally more than 6.5 m.
The following steps are involved in temporary repairs.
1. Two paint marks are made on either side of the fractured
(97)
joint at a distance, say X and Y as shown in the Fig 6.1.
2. These distances X and Y are correlated to the length of
the closure rail piece which is to be inserted into the track.
Let the length of this closure rail piece be L. While
replacing the fractured rail, the principle to be observed is
that the rail inserted into the track should be equal to the
length of the rail removed.
Here length of rail inserted = L + 2 welds (50mm)
Length of rail removed = X + Y + 2 saw cuts (say 1mm )
Hence L + 2 welds (50 mm) = X + Y + 2 saw cuts (say
1mm)
This relationship enables fixing up of paint marks on either
side of the fractured rail joint at distances of X and Y from
the fracture location.
3. During a block of adequate duration, rail cuts are made on
either side of the fractured joint at the paint marks made
earlier. The closure rail piece of length ‘L’ is inserted into
the gap created and fishplates with screw clamps fixed at
the two joints.
Fig 6.1
(98)
(99)
4. In the same block if time permits, or in another block one
of the gaps is adjusted to 25 mm (for SKV welding) and
welding performed.
If the other gap is also 25 mm then the other joint could
also be welded. However, if the other gap is not 25 mm, as
will generally be the case, then the other gap is fishplated
with screw clamps and opened to traffic.
5. For welding the other gap, a tensor is used. It will be used
for reducing the gap to 25 mm. This is done by removing
sleeper fastenings over a given length and applying the
requisite pull by tensor. With the tensor in position and
gripping the rail web, the rail joint is welded. The tensor is
kept in position till weld metal cools down. After the weld
metal has cooled down, the tensor is removed, and
fastenings removed over a length of 125 m on either side
of the weld. The rail is tapped to equalise the stresses and
fastenings put back in position.
6. To summarise, the process of temporary repairs is carried
out without adding any additional rail metal to the LWR. If
the gap at fracture is bridged by providing a longer length
closure rail, there will be a drop in the neutral temperature,
creating high compressive stress in the rail during the
summer season. The temporary repairs should be carried
out in the supervision of a PW Mistry/PWI.
7. A new development in welding technology is wide gap
welding. This enables fracture repairs to be done by
providing a single weld instead of two welds as was done
earlier. The method of repairs has been indicated in Fig.
6.2.
A,B are paint marks on either side of fracture.
Rail inserted = 75mm. (wide gap weld)
Rail removed = X+Y+2 saw cuts (1mm)
75 = X+Y+2 saw cuts
C. Permanent Repairs :
This will involve destressing the entire LWR after a
number of fractures have occurred in the LWR. The
Manual prescribes destressing to be done when the
number of fractures exceeds 3 per km.
D. Equipments required for fracture repairs :
1. Fishplates/ Joggled fishplates with bolted clamps.
2. One metre long fishplates with bolted clamps.
3. Rail Closure Pieces with different lengths.
4. Welding equipment with all accessories.
5. Tensor for obtaining the standard gap.
6.3 BUCKLING :
6.3.1 Buckling is the phenomena describing the sudden lateral
shift in the alignment of an LWR to relieve the built–up
compressive forces during the summer months as the temp-
erature rises above the destressing temperature. Buckling
results in complete distortion of the track geometry affecting
safety and it is not possible to pass a train over the buckled
track. Over the years, lateral stability of LWR track in hot
weather conditions has been a source of great concern to track
engineers. Initially it was thought that the slender rail section
would not be able to take the high compressive forces gene-
rated during the summer season. Subsequent investigations by
(100)
Fig. 6.2
LOCATION OF FRACTURE
various railways, however, indicate that the track strength
against buckling was contributed by not only the rails, but the rail
sleeper assembly with fastenings, and the ballast contributed in
a substantial measure to the strength against buckling.
6.3.2 Some of the factors which could lead to buckling are :
1. Non–observance of the specified temperature
restrictions while performing maintenance operations in
an LWR.
2. Lack of ballast affecting the lateral and longitudinal
ballast resistances.
3. Missing fittings.
4. Settling formation resulting in poor alignment of track.
5. Improper functioning of the SEJ.
6.3.3 Steps to be taken to avoid buckling :
1. Ensuring proper ballast profile.
2. Full complement of fastenings and anchors.
3. Observance of specified temperature restrictions (t
p
< t
d
+
10
0
C) during maintenance operations.
4. Introduction of hot weather patrolling when the prevailing
rail temperature goes beyond t
d
+ 20
0
C, (t
p
>t
d
+20
0
C).
5. Controlling misalignments in track.
6. Keeping a close watch on SEJ gaps specially during
extreme temperatures.
7. Proper repairs of fractures i.e. avoiding addition of metal at
the fracture location during repairs.
6.3.4 Steps to be taken in face of impending buckling :
On detecting severe sunkinks or noticing hollowness of sleepers as
detected by a canne–boule, the following steps are suggested :
1. The section to be blocked or speed restrictions to be
imposed, depending upon the severity of the situation.
2. Additional ballast to be dumped on the shoulder, by taking
out, if required, ballast from the centre of the track.
(101)
3. At tight rail locations, rail should be cut out from the track.
In the morning times, this could be achieved using a
hacksaw blade. However, this will not be possible when the
rail is under compression as it will tend to pinch the blade.
Gas cutting to cut rail out will have to be resorted to.
Subsequently the heat affected martensite zones could be
removed by cutting three inches on either side of the gas
cut by a hacksaw blade. The golden principle to be
followed is : “When in doubt, cut rail out.”
6.3.5 Repairs to be undertaken in the event of buckling :
1. Each case of buckling shall be investigated by the AEN
soon after its occurrence and a detailed report submitted to
the DEN/Sr DEN.
2. The rectification shall normally be carried out in the
following stages under supervision of the PWI.
(A) Emergency Repairs
1. This repair is carried out to restore traffic on the
section. A 6.5 m long rail piece will be cut out from
the buckled track at the location of buckling,
resorting to gas–cutting if required.
2. After removal of the rail piece, it will be possible to
slew the track back into proper alignment.
3. A closure rail piece of suitable length could now be
inserted into the track and section restored after
fixing fishplates and screw–clamps.
(B) Permanent Repairs
The clamped closure rail piece will be welded at either end.
To get the required gaps for welding, rail cutting equipment
will be required. In order to complete the repairs,
destressing of the entire LWR will be carried out as early as
possible.
6.4 Inspection of the LWR and Record keeping :
6.4.1 Inspection: While an LWR section, reduces the maintenance
requirements it necessitates intensive inspections at supervisory
and officers' level. The Sr. DEN / DEN, AEN & PWI and other
inspecting officials should pay special attention to the aspects
given below while inspecting LWR sections.
(102)
1. Ballast adequacy and maintenance of stipulated
ballast profile specially at locations where the profile is
likely to be disturbed due to trespass.
2. Special attention at vulnerable locations such as
curves, level crossings, girder bridge approaches, etc.
3. Knowledge of rules and regulations (specially
temperature restrictions) of Mates, Keymen, Gangmen
and P Way Mistries for track maintenance in LWR
territory.
4. Action to be taken by P. Way Mistry, Mate, Keyman
and Gangman in the event of a fracture or buckle.
5. Ultrasonic Flaw Detection of rails and welds should not
be in arrears.
6,. Inspection of SEJ gaps and creep movement in
central portion of LWR / CWR should be as per
schedule given below:
(i) PWI / APWI - To measure the SEJ gaps
alternately once in 15 days during the two hottest and
two coldest months of the year. In the remaining 8
months, they will measure at 2 monthly intervals again
alternately.
(ii) The sectional AEN will measure the SEJ gaps once in
6 months preferably during the coldest and hottest
months.
6.4.2 Records:
(1) The PWI should maintain a permanent register called
the LWR section register. This register should record
various details of the LWR as laid down in annexure XI
and XII of the LWR Manual.
(2) An indication plate should be fixed on the cess at each
SEJ, showing the date of destressing, destressing
(103)
temperature and length of LWR / CWR.
(3) Inspection of SEJ gaps and creep movement in the
central portion will be recorded as per laid down
frequency in the proforma prescribed in Annexure-
XIII(A) and XIII(B).
(4) PWI / AEN / DEN will carefully study the SEJ gaps
and creep in the central portion before deciding the
remedial measures required to be taken.
(5) AEN will analyse the observation of each LWR / CWR
in his jurisdiction and give a certificate at the end of
the LWR / CWR section register before onset of
summer regarding satisfactory behaviour of all LWRs /
CWRs on his section. DEN / Sr.DEN will scrutinize
observations of each LWR /CWR, initial each page
and send exception report to Territorial Chief Engineer
for his decision / orders.
6.5 Duties, responsibilites and staff training
Readers may refer to para 9 of LWR Manual 1996 to know
more about the duties, responsibilites and training of staff for
working in LWR territory.
(104)
CHAPTER VII
SPECIAL TOPICS
7.1 Buckling Phenomena : As described earlier, buckling is the
sudden lateral shift in the track alignment to release the built up
compressive forces in the rail. The strength of track against
buckling or what is described as lateral stability of track has been
investigated in great detail by various railways. The studies
conducted by various railways and the results thereof have been
discussed in this chapter.
7.2 Tests by German Railways: Results of a series of track
buckling tests conducted for the Federal German Railways were
reported by F. Birmann and F. Raab in 1960. The test facility was
located at the Technical University of Karlsruhe. The track section
was 46.50m and was confined at both ends by reinforced concrete
blocks. The following results were obtained from the tests :
1) In all the tests the track buckled laterally. The buckling
modes exhibited 2,3 or 4 noticeable half waves each of
length 5 to 6 metres. The largest amplitude of
displacements was about 25 centimetres. This implied that
a buckled track could have several shapes with buckling
taking place in several wave forms.(Fig 7.1) Buckling in the
form of a ‘C’ could occur a on sharp curve (First wave
form) while buckled track resembling an ‘S’ shaped curve
is generally evidenced on straight tracks (2
nd
wave form).
The force diagram after a buckle is shown in Fig 7.2. It
indicates that while a track physically buckles over a length
‘l’ the force diagram is affected over a length ‘a’ where ‘a’
is several times ‘l’.
2) Straight tracks with smaller lateral imperfections
buckled at much higher temperature increases than those
tracks with noticeable lateral imperfections. Buckling of
straight tracks occurred suddenly with a loud bang
(explosive buckling) while the imperfect track buckled
gradually and quietly(passive buckling).
(105)
Fig. 7.1
First waveform
Second waveform
Third waveform
(106)
(107)
3) With use of different fasteners, the buckling load varied by
as much as 25%
4) Over a period of time with reversal of temperatures there is
an accumulation of undesirable permanent lateral track
deformations for temperature increases which do not cause
actual track buckling but definitely increase the
buckleproneness. This is shown in Fig. 7.3.
7.3 Studies Conducted by British Transport Commission : In order
to study the conditions and factors affecting the stability of the
Long Welded Rails a large testing program was started in 1953 by
the Civil Engineering Laboratory of the Western Region of British
Railways. These researches were carried out and described by Mr
D.L.Bartlett, Assistant Director of Research (Engineering),
Research Department, British Railways.
Fig. 7.2
Length over which force diagram
Length of track where track has
has been affected 'a'
physically gone out of alignment
'l'
Reduced compressive force
q
Fig. 7.3 ACCUMULATION OF LATERAL DISPLACEMENTS
(108)
T4 & T1 are temperatures at which
significant lateral displacements
occur
T5 & T2 are track buckling
temperatures
1
.

1
2
0

f
t

l
e
n
g
t
h

o
f

t
r
a
c
k
2
.

E
n
d

a
n
c
h
o
r
a
g
e

b
l
o
c
k
s
3
.


T
i
e

b
a
r
s
4
.

R
e
s
t
r
a
i
n
i
n
g

b
e
a
m
s
5
.

H
y
d
r
a
u
l
i
c

j
a
c
k
s
6
.


E
l
e
c
t
r
i
c

h
e
a
t
e
r
s
7
.

D
i
a
l

g
a
u
g
e
s

r
e
g
i
s
t
e
r
i
n
g

l
o
n
g
i
t
u
d
i
n
a
l

r
a
i
l

m
o
v
e
m
e
n
t
8
.

T
h
e
r
m
o
m
e
t
e
r
s
9
.

D
i
a
l

g
a
u
g
e
s

r
e
g
i
s
t
e
r
i
n
g

l
a
t
e
r
a
l

r
a
i
l

m
o
v
e
m
e
n
t
F
i
g
.

7
.
4
1
(109)
7.3.1 Test Arrangement (Fig. 7.4)
The main tests devised for the purpose of carrying out buckling
tests was a 120 feet test bed upon which could be built, complete
in every respect a length of track, the whole capable of being
subject to thermal stresses. The arrangement of the test bed was
such as to simulate the central portion of a long welded rail length
on site which does not move longitudinally with temperature
change. The test bed was laid inside a disused tunnel where a
constant ambient temperature could be expected.
The 120 feet track rails were anchored at each end to concrete
blocks sunk below ground level. This was sufficient to prevent
rotation of the track and change of gauge but not to prevent the
expansion of the rails. The latter was controlled by four tie bars,
two on each side of and clear of the test track. Any tendency for
the rails to expand could be counteracted by the jacks, although it
must be stressed that the jacks were not directly used to induce
compression in the rails. Four dial gauges altached to an
independent datum registered any longitudinal movement of each
rail end during the tests. By operating the jacks the rail lengths
could be kept sufficiently close to their original values to be
consistent with actual conditions in the field.
Heaters : Electric heaters with parabolic refiectors were used to
simulate the heat radiation from the sun; they were situated on one
side of each rail at a distance determined experimentally so that
the rate of heating was not excessive.
Thermometer : Normal glass and mercury thermometers inserted
in sockets drilled mainly in the head of the rail were used to
measure the temperature.
Misalignment : This is the offset of the rail from the straight. The
length of misalignment is the length over which misalignment
occurs. The track was laid initially as straight as possible and then
given a small misalignment over a given length.
Methodology of Test
Using the above setup, the longitudinal load required to buckle the
track was determined experimentally for different types of sleepers,
(110)
(111)
fastenings and ballast packing conditions. Using theoretical
methods the longitudinal load required to buckle a track was
determined and the same compared with experimental values.
7.3.2 Buckling Load Formula:
The formula derived for the longitudinal load required to buckle a
straight track is :
q
L
D
C
L
EIs
P
Π Π
+
Π
·
16
2
2
2
+
q
L W
2
2
max
Π
Where
Is is the moment of inertia of the two rails put together in the
horizontal plane.
L is the distance between the points of contraflexure of the buckled
track.
C is the torsional coefficient for the given type of fastening
α C T ·
, Where T is the torque resisting buckling and
α is the angle of twist for the fastening due to rotation of the rail on
the rail seat
D is sleeper spacing
q is the misalignment of the track over length L
if Wmax is the lateral ballast resistance per meter length of track
and W is the lateral ballast resistance per sleeper then
Wmax =W/D
Analysing the above expression, it can be seen that :
1)
2
2
L
WIs Π
represents the contribution of the rails to resistance
against buckling. Little can be done to this term, as it is dependent
mainly on the properties of the rail.
2)
q
L
D
C Π Π
16
2
represents the contribution of the sleeper/fastening
combination to the resistance against buckling. Here clearly a
reduction in sleeper spacing D or an increase in the fastening
torsional co-effecient C will cause an increase in the overall
resistance to buckling
3)
q
L W
2
2
max
Π
reperesents the contribution of the lateral ballast
resistance.
The following points are to be noted:
1. If the track were perfectly straight and points of equal load
application central for each rail, then the track would not buckle
however great the longitudinal compressive force. However, in
practice no track exists under these ideal conditions and a
misalignment of ‘q’ over a length ‘L’ will always be present. In any
case, it is evident that the lower the L /q ratio, the smaller will be
the buckling load. It means that large misalignments significantly
reduce the strength against buckling.
2. Experimentally it has been observed that when a buckling
occurs, the sleepers remain at right angles to the original track
alignment. For this to occur, the rail must rotate on the rail seat.
Clearly, only one thing resists such a rotational movement and this
is the torsional resistance (denoted by torsional co-efficient C)
afforded by the fastenings. Clearly the buckling load is
proportional to torsional resistance.
3. ‘L’ the length of buckled track is taken as 20 feet for all cases.
In actual fact for a given combination of C,D, Wmax and q there
exists only one value of ‘L’ which will yield a minimum value of ‘P’
(the buckling load). Hence for various combinations of these
variables, a range of ‘L’ values would emerge. For practical use
however, ‘L’ is chosen as 20 feet and the value of ‘q’ as 1/4
inch(6mm).
4. The relative contributions of rails, rail sleeper fastenings. and
ballast would depend upon the actual conditions prevailing at site.
Under normal conditions the percentage contributions could be
10%,30% and 60% respectively.
5. The buckling load values as determined experimertally show a
fair correspondance (within a few per cent) with the values
determined from theoretical calculations.
(112)
6.A PWI can ensure that the track remains safe against buckling
by:
1. Reducing the lateral misalignment in the track.
2. Ensuring that no sleeper rail fastenings are missing.
3. Providing full complement of ballast in the track as per precribed
ballast profile.
7.4 Static Buckling and Dynamic Buckling
The discussion so far has been centred on buckling caused by
longitudinal compressive force buildup due to rise of temperature
above the stress-free temperature. This buckling due to thermal
loads alone is called static buckling. The industry today is more
concerned with buckling caused by the movement of a train on the
track in the presence of thermal loads. Such a buckling is called
dynamic buckling. The effects of a moving train which could
contribute to dynamic buckling are as given below :
1) Loaded axles of a moving train cause the track to be lifted in front
of, in the rear of or even between the moving axles. The wave so
created as seen in the vertical profile of the rail in front of the engine
is called the precession wave,in the rear the recession wave and in
between the axles, the central wave. Any of these waves
Fig. 7.5 (a)
(113)
(114)
A
S
S
U
M
E
D

L
A
T
E
R
A
L

B
U
C
K
L
I
N
G

M
O
D
E
F
i
g
.

7
.
5

(
b
)
I
N
I
T
I
A
L

L
A
T
E
R
A
L
I
M
P
E
R
F
E
C
T
I
O
N
R
E
C
E
S
S
I
O
N

W
A
V
E
D
I
R
E
C
T
I
O
N

O
F
T
R
A
V
E
L
q
P
R
E
C
E
S
S
I
O
N

W
A
V
E
q
C
E
N
T
R
A
L

W
A
V
E
C
E
N
T
R
A
L

W
A
V
E
(115)
could be critical enough to cause loss of contact between the
ballast and the sleeper soffit resuting in the loss of lateral ballast
resistance thereby making the track buckleprone (Fig7.5(a)&(b)).
2) Tractive and braking forces applied by the moving train
change the force level in the LWR and continuous braking
at a given location could result in buildup of compressive
forces creating buckling tendencies in the rail.
3) The hunting motion of the moving train over lateral
mislignments in the track could create large lateral forces
producing buckling tendencies.
4) Vibrations induced by the moving train could disturb the
ballast and lower the lateral ballast resistance.
7.5 Dynamic Track Buckling Model:
The effect of a moving train increasing tendency of a track to
buckle when the temperatures are rising or the response of the
track to disturbing lateral forces is depicted by what is called
DYNAMIC TRACK BUCKLING MODEL. This model is essentially a
relationship between the lateral track displacement and the
temperature increase over the force free or neutral temperature.
The model is depicted in the figure (Fig7.6).
The model has 3 limbs as shown: AB is the prebuckling limb while
B
D
1
2 3
T
BMAX
T
BMIN
T
e
m
p

R
i
s
e

a
b
o
v
e

T
N
T
N
A
C
Displacement
Fig. 7.6 DYNAMIC TRACK BUCKLING MODEL
BC and CD are post-buckling limbs. At B when the temp. rises to
T
BMAX’
the track becomes unstable where even an infinitesimal
lateral force will cause the track to buckle. Below temp T
BMIN
at
point C, even a large force will not be able to buckle the track.
Between points ‘B’ and ‘C’ a moving train could impart sufficient
force to buckle the track. Between ‘B’ and 'C' the track on buckling
will first move to an unstable buckled phase on curve BC and
subsequently to a stable buckled phase on CD. It is assumed that
if the track can be brought into position 2 it will automatically move
into position 3. At T
BMAX,
the energy required to buckle a track is
almost zero while below T
BMIN
the energy required to buckle a track
is much larger than that which could be provided by a moving
train. Between T
BMAX
and T
BMIN,
the transition from the pre-buckling
stage to the unstable buckled state and to the stable buckled state
could be effected under the influence of energy imparted by the
moving vehicle.
Various softwares have been developed to predict the T
BMAX
and
T
BMIN
temperatures for given track and rolling stock parameters. In
the USA, the program developed is called CWR-BUCKLE. Other
software programs are CWRSAFE and related programs. The
inputs to these programs are :
1) Rail section
2) Track curvature
3) Rolling stock characteristics
4) Lateral ballast resistance
5) The misalignment in the track.
Using these programs the T
BMIN
and T
BMAX
of the given track for a
given set of parameters are determimed. The policies regarding
LWR maintenance can then be decided. These could include :
1) Allowable Temperature rise above t
n
for maintenance
activities
2) Temperature at which track enters the danger zone, and
necessitates hot weather patrolling.
7.6 CWR Safety Assurance Program:
The discussion above could form the basis of a continuous welded
track buckling safety assurance program. For safe operations of
CWR track with respect to buckling, the allowable temperature
increase T
ALL
over the neutral temperature should be greater than the
(116)
difference between the maximum anticipated rail temperature T
MAX
on a given day and the neutral or stress-free temperature T
n
i.e.
T
ALL
> (T
MAX
-T
n
)
The expression on the right is the anticipated rail temperature rise
over T
n
while T
ALL
is the allowable temperature rise which could be
determined from the values of T
BMIN
and T
BMAX..
T
ALL
will be
somewhere between these two extreme temperature values
depending upon the track parameters,level of maintenance and
monitoring and the degree of risk the railway administration is
willing to take. A conservative approach would be to fix the T
ALL
at
T
BMIN
value. However, a better approach would be to fix T
ALL
higher
than T
BMIN
if the railway has good track maintenance and
monitoring procedures in place.
The expression given above indicates that for a safe CWR
assurance program two temperatures need to be determined:
(i) T
ALL
which is the allowable temperature rise above the
neutral temperature for a given set of track and vehicle
parameters.The single most significant factor which will
govern T
ALL
for a given set of track and rolling stock
parameters is the lateral ballast resistance. The
ralationship between T
ALL
and the lateral ballast resistance
will be in the form of a graph. This could be given to the
field maintenance engineer to enable him to predict the
allowable temperature rise over the neutral temperature for
a given value of the lateral ballast resistance. (Fig 7.7)
(ii) The neutral temperature or the stress free temperature of
the track.
7.7 Field Determination of the lateral ballast resistance:
A convenient metod to determine the lateral ballast resistance per
sleeper has been developed in the USA.It is called the single tie
push test (STPT).
Test Methodology: The rail is freed from the sleeper by removing
the fastenings and using a hydraulic jacking equipment the tie is
pushed transversely to the track. With load transducers and
gauges, the loads and corresponding displacements are recorded.
The plot gives the maximum lateral resistance of ballast. For getting
the average value the test could be performed on 3 sleepers over a
50 feet length. Once the lateral ballast resistance value is obtained
(117)
(118)
graph could be given to the field maintenance engineer to enable
him to predict the allowable temperature rise T
ALL
over the neutral
temperature.
7.8 Neutral Temperature, Its Variation and Determination
7.8.1 Introduction :
The neutral temprature t
n
or the stress-free temperature t
o
is the rail
temperature at which the rail is free of longitudinal stress, or
longitudinal stress is zero. Till now the assumption was that once
an LWR was destressed at temperature t
d,
, t
d,
was the stress free
temperature of the rail. However, there is experimental evidence to
indicate that the above assumption is not correct and the stress-
free temperature of the rail tends to shift away from the
destressing temperature. Accurate determination of the rail stress
free temperature is of vital importance, because it is this
temperature which determines the force level in the LWR
Fig. 7.7 Relationship between T
ALL
and lateral
ballast resistance.
(119)
[ ]
n P
t t AE P

· α
The above expression also indicates that if due to any reason the
value of t
n
were to fall, it would automatically increase the
compressive force in the rail and beyond a certain level could
cause the track to buckle. Another way of putting it is that a
change in the rail neutral temperature is equivalent to changing the
force level in the rails for the same values of t
P .
7.8.2 Factors which could cause a shift in the Rail Neutral
Temperature:
1) Movement of the rail in the longitudinal, lateral and vertical
directions : If the CWR were to be fully constrained, then there
would be no change in the neutral temperature. Since the rails
cannot be fully constrained in all directions, elongation or
contraction can occur whenever the track is subject to train and
environmentally induced loads. Railway track motions relevant to t
n
variation occur in the following three basic kinematic modes :
(i) Rail longitudinal movement
(ii) Track lateral shift
(iii) Track vertical settlement.
Consider a CWR being laid at Temperature t
L
and there is no rail
longitudinal force at this temperature. Assume that the rail
displacements (u,v and w in the longitudinal, lateral and vertical
directions respectively) are measured with respect to an initial
equilibrium configuration when the rail temperature is t
L.
These
displacements may be due to a number of causes, and in many
cases are not recoverable due to the inelastic nature of the ballast.
From the displacements, the longitudional strain εx in the rail at
any given temperature t
P
can be calculated from the fundamental
equations of theoretical mechanics.
,
[ ]
L P
t t
x x
v
x
u
x − +
1
1
]
1

¸

,
_

¸
¸


+
,
_

¸
¸


+


− · α
ϖ
ε
2 2
2
1
2
1
[ ]
L P
t t − α is a compressive strain taken as positive in the
longitudinal ‘x’ direction.
The force in the CWR at t
P
in the longitudinal ‘x’ direction will be
(120)
P=AE εx
1
1
]
1

¸

¹
¹
¹
;
¹
¹
¹
¹
'
¹

,
_

¸
¸


+

,
_

¸
¸


+


+ − ·
2 2
2
1
2
1 1
x x
v
x
u
t t AE
L P
ϖ
α
α
-------1
If t
n
is the neutral temperature of the CWR then
[ ]
n P
t t AE P − · α -------------2
Comparing equations 1 and 2 it is quite evident that
¹
¹
¹
;
¹
¹
¹
¹
'
¹

,
_

¸
¸


+
,
_

¸
¸


+
,
_

¸
¸


+ ·
2 2
2
1
2
1 1
x x
v
x
u
t t
L n
ϖ
α
where
x
u


,
x
v


, and
x ∂
∂ϖ
are all tensile strains.
If the movement of the rail leads to additional tensile stress, then
the neutral temperature will increase beyond t
L
. On the other
hand, if u, v and w the rail displacements cause compressive
strains, they will cause t
n
to drop below t
L .
On a curve of radius ‘R’ if the track is shifted by an amount equal
to ‘V’

then

¹
¹
¹
;
¹
¹
¹
¹
'
¹

,
_

¸
¸


+
,
_

¸
¸


+ +
,
_

¸
¸


+ ·
2 2
2
1
2
1 1
x x
v
R
V
x
u
t t
L n
ϖ
α
Movement of the rail in the x, y and z directions may be due to
the following reasons :
i) Rail Longitudinal Movement – This may occur due to
train action (braking and accleration) or due to wheel
rolling action.
ii) Track Lateral Shift – This may occur due to bogie
hunting action, by wheels negotiating a lateral
imperfection or in case of curves, by vehicles operating
in excess of or below the balance speed. For tangent
track, the effect of lateral movement on the neutral
temperature shift is likely to be small. For curved
track, however, the shift can be significant.
iii) Track vertical Settlement – Vertical wheel loads can
induce differential settlement of the ballast which
would cause development of longitudinal strains in
CWR specially for new or recently surfaced tracks.
,
Apart from these displacements which could cause a shift in the
rail neutral temperature, two more factors could cause a shift.
These are :
1. Rail/Track Maintenance – Track maintenance
activities involving lining, lifting, removal or
application of rail anchors, trackline repairs all
will cause a shift in neutral temperature.
2. Rail “Rolling Out” – Due to vehicle loads, plastic
deformation occurs in the top layers of the rail
head. Experiments conducted by the British
Rail showed rolling out of rail is pronounced in
first 3 months after laying new rails and
continues for about an year. The basic
mechanism involved is that the rolling contact
loads change the residual tensile stresses in
the top layer of new rails into compressive
stresses. The rail residual stresses will affect
the neutral temperature of the rail (Fig 7.8).
7.8.3 Neutral Temperature Measurement
7.8.3.1 A satisfactory neutral temperature measuring device
should satisfy the following fundamental requirements :
1. The measuring instrument should be portable and not
permanently attached to the rails.
2. The instrument should give absolute values and not
relative values. Site specific calibration should not be
involved.
(121)
3. The technique should be independent of longitudinal
residual stresses in rail. The residual stresses are not
associated with the rail longitudinal force since they are
self-equilibrating in the sense that their resultant force and
moment are zero. As a result, any technique which relies
on measurement of local stresses for the longitudinal force
can have large errors.
4. The technique should be non-destructive.
5. The technique should be fairly accurate with
measurements within +1
0
C.
Fig. 7.8 Rail residual stresses
(122)
(123)
7.8.3.2 Of the number of techniques available for neutral
temperature measurement, the following can be considered as
reasonably developed :
l Berry Gauge – Simple mechanical gauge to
measure change in length.
l British Rail Vibrating Wire – Measures the rail
force as a function of the frequency of a wire
vibrating in a hole in the rail web.
·l Strain Gauge –Uses a four arm Wheatstone
bridge to measure the rail strains.
Techniques under Research –
1. Flexural wave propagation
2. X-ray defraction
3. Accousto-elastic
4. Magnetic coercion
5. Barkhausen Noise – This principle is being used in Rail
Scan Equipment.
6. Electromagnetic Accoustic Transducers
7. Laser ‘Spackle’
7.8.3.3 Rail Uplift Method :
A new approach based on rail beam column response has shown
considerable promise. It is based upon the fact that if the rail can be
held at two points at some distance apart and a concentrated load
applied at the centre of this portion, the rail behaves like a beam
column and its deflection is influenced measurably by the
longitudinal load in the rail. Clearly a compressive longitudinal load
will increase its deflection, whereas a tensile load will reduce it.
Besides the longitudinal force, the deflection is dependent on the
rail flexural rigidity, EI, applied load Q and the nature of the end
constraints. It is necessary to design a rig such that for all
locations and measurements, the end conditions are sufficiently
repeatable. As far as the end conditions are concerned, they
depend upon the nature of constraint provided by the rig.
Generally, the conditions are elastic supports (in between pure
simple supports and completely fixed supports). Fixed support
conditions improve the sensitivity, but need large applied loads.
Repeatability of the end conditions is an important consideration
for successful application of the technique.
The deflection

is given by
) 1 (
1
1
3
− − − − − − −

· ∆
c
P
C
EI
QL
λ
where C is the longitudinal compressive force in the rail
Q = Verticle load applied at centre of rail.
λ
= Numerical constant value depending on the end conditions,
Pc = Critical buckling load for the beam column of length 2L for
the specific end conditions.
The first factor in the above equation represents the deflection
under the concentrated load in the absence of any longitudinal rail
force. The second factor is the magnification factor due to
longitudinal force.
The above equation shows that for a given value of C (rail
longitudinal force), Q and

are proportional to each other. This
is depicted in the given figures (Fig 7.9 & Fig 7.10)
7.8.3.4 VERSE METHOD
Practical use of this principle has been made in the technique
called ‘VERSE’ developed by VORTOK International, UK and AEA
Technology Rail. The equipment comprises of a frame featuring a
hydraulic lifting device, a load transducer and a displacement
(124)
F
i
g
.

7
.
9

R
a
i
l

U
p
l
i
f
t

M
e
t
h
o
d
.
R
A
I
L

U
P
L
I
F
T

D
R
I
V
E
(
R
U
D
)
R
E
C
O
R
D
E
R

A
N
D

P
L
O
T
T
E
R
P
O
W
E
R

P
A
C
K
(125)
transducer. The measurement systems are connected to a rugged
handheld computer.
The rail must be in tension at the time of measuring the stress free
temperature (SFT). Taking measurements requires around 30 m of
rail to be unclipped and placing rail support spacers at 10 m on
either side of the measuring point (Fig 7.11). A maximum force of
one tonne is applied and the load and displacements measured by
the transducers relayed to the handheld computer. The measured
data along with some other data such as ambient rail temperature,
rail profile and height of rail is fed into the computer to obtain the
SFT result. The height of the rail is included to take account of the
T = C = 0
C = 25 TONS
C = 50 TONS
C = 75 TONS
C = 100 TONS
T

=

2
5

T
O
N
S
T

=

5
0

T
O
N
S
T

=

7
5

T
O
N
S
T

=

1
0
0

T
O
N
S
Fig. 7.10 Graphical Method for neutral temperature
determination.
Q
(126)
T= tensile force
C= compressive force
Q= centrally applied
load
= deflection at
centre
D
D
(127)
F
i
g
.

7
.
1
1

V
E
R
S
E

M
E
T
H
O
D

F
O
R

N
E
U
T
R
A
L

T
E
M
P
E
R
A
T
U
R
E

D
E
T
E
R
M
I
N
A
T
I
O
N
.
rail head wear and rail grinding which will naturally affect the
stiffness of the rail. Validation of VERSE technique has been
carried out by AEA Technology, one of Britain’s leading
technology companies.
(128)
List of References:
1. Report of Committee on Welded Rails – Civil Design
Directorate Report No.1.
2. Experiments on the stability of Long Welded Rails –
British Transport Commission London 1961.
3. Study of Rail Temperature in India, RDSO/C-146.
4. Longitudinal Ballast Resistance of Different Types of
Sleepers Broad Gauge and Meter Gauge, RDSO/C-
148.
5. Lateral Ballast Resistance of Different Types of
Sleepers, RDSO/C-156.
6. Long Welded Rails on Girder Bridges: RDSO/C-166.
7. LWR on Girder Bridges, RDSO/C-169.
8. LWR on Girder Bridges with Creep Resistant
Fastenings: RDSO/C-170.
9. UIC Code 774-3R. Track/bridge Interaction
Recommendations for calculations
10. Modern Permanent Way – By M. Srinivasan.
11. Manual of Instructions on Long Welded Rails 1996.
12. ‘‘ The Neutral Temperature Variation of Continuous
Welded Rails’’ By: A. Kish, G. Samavedam and D.
Jeong
(129)
13. ‘‘How UP Achieves Lateral Track Stability on CWR’’
By William C. Thompson.
14. ‘‘ A New Safety Philosophy for CWR ’’
By Coenraad Esveld.
(130)

(ii)

PREFACE

Long Welded Rail (LWR) has now become synonymous with modern track structure with a major portion of Indian Railways track having long welded rails. It is imperative that permanent way men understand all its facets , be it welding, laying or maintenance so that full benefits are reaped . With this objective, IRICEN publication on LWR was printed in 1988 which of course requires revision. This publication is an updated version with a completely new look incorporating the latest correction slips and provisions of the LWR Manual. The publication highlights the evolution of the LWR over the years with brief references to the research work carried out in RDSO and foreign railways on various aspects of the LWR. A brief description of the various SEJ layouts now available, latest provision of LWR on bridges with comments on the state of art, neutral temperature and its measurement are also included. It is hoped that this publication will go a long way in helping track engineer to understand the intricacies involved in laying and maintaining LWR track. This book has been authored by Shri Ajit Pandit, Sr. Professor & Dean of this Institute. If there are any suggestions or discrepancies, kindly write to the undersigned.

Shiv Kumar Director IRICEN

(iii)

ACKNOWLEDGEMENTS

While covering the subject on Long Welded Rails at IRICEN the absence of an updated publication on the subject covering the state of art and latest instructions was acutely felt. The publication printed in 1988 required revision to incorporate the provisions of the LWR Manual 1996, including the latest correction slips. This IRICEN publication is a result of the desire to fill the gap and produce a documentation which would be useful for all practicing civil engineers on Indian Railways. It would be appropriate to mention the support and assistance rendered by IRICEN faculty and staff in preparing this publication. Special mention may be made of Shri Sunil Pophale, Head Draftsman who rendered valuable assistance in preparing the drawings. Shri Dhumal, PA assisted in editing the manuscript. Shri R.K. Verma, Senior Professor/Track gave valuable suggestions from time to time. Above all, the author is grateful to Shri Shiv Kumar, Director/IRICEN for his encouragement and guidance for preparing the document.

Ajit Pandit Senior Professor & Dean

1 3.4 Evolution of Long Welded Rail Some Basic Definitions An Explanatory Note on the Short Welded Rail Advantages of Long Welded Rail PAGE NO.3 3.2 3.3 4.4 4.2 2.3 2.1.2 4.5 2.4 Estimation of Thermal Movements Switch Expansion Joints Phenomenon of Hysteresis Gap measurements at Switch Expansion Joints 28-46 CHAPTER IV : PERMITTED LOCATIONS AND TRACK STRUCTURE 4.1 2.5 General Considerations Alignment Gradients Track structure LWR on bridges 47-77 . 1-8 CHAPTER II : PRINCIPLES OF LONG WELDED RAIL 2.4 2.2 1. 1.6 Basic Principles Force Diagram Importance of rail temperature Breathing Length Longitudinal Ballast Resistance Lateral Ballast Resistance 9-27 CHAPTER III : THERMAL MOVEMENTS AND HYSTERESIS 3.3 1.1 4.(iv) INDEX/CONTENTS CHAPTER I : INTRODUCTION TO LONG WELDED RAILS 1.

5 Introduction Fractures Buckling Inspection and Record keeping Duties.3 Regular Maintenance operations 5. 104 INSPECTION & RECORD KEEPING 6.1 7.2 7.5 7.2 Destressing 5. its variation and determination 129 List of References .7 7.4 7.5.1 Laying of an LWR 5.1 6.(v) CHAPTER V : NO.8 Buckling Phenomena Tests by German Railways Studies conducted by British Transport Commission Static and Dynamic Buckling Dynamic Track Buckling Model CWR Safety Assurance Program Field determination of lateral ballast resistance Neutral Temperature.6 7.3 7.3 6.4 6. responsibilities and training 96- CHAPTER VII : 105-128 SPECIAL TOPICS 7.2 6. LAYING AND MAINTENANCE 5. Deep screening in LWR territory PAGE 78-95 CHAPTER VI : UNUSUAL OCCURENCES IN LWR.4 Special Track Maintenance 5.

.

Rails of 65m and 78m and even longer rails upto 480m are being planned for manufacture in the near future in the steel plants across the country. 1.To achieve this various innovative techniques have been used over the years. Although welding of rails was started in 1905 itself. the length of the rail was governed by the length of cooling boxes in the rail manufacturing steel plant.2 At the beginning of the last century the standard length of rail was generally 12m. Subsequent advancements in the manufacturing process have enabled rolling of longer rails possible. Apart from the logistic considerations of rail transportation and its loading and unloading. 1. Two ways to achieve this are: (1) Increasing the length of rails by rolling longer rails. as controlled cooling after the rolling process was necessary.(1) CHAPTER .1. 26m long rails are now being manufactured by SAIL in the Bhilai Steel Plant. Jindal Steel & Power Ltd are setting up a steel plant in Raigad dist. (2) Welding the rail joints.I INTRODUCTION TO LONG WELDED RAILS 1.1.1 The dream of a jointless track has fascinated track engineers ever since the first railway was laid and efforts are continuing even today for achieving a fully continuous track without joints.1.1 Evolution of Long Welded Rails 1. where longer length rails are likely to be manufactured.3 . of Chhattisgarh state. commercial welding on any considerable scale became common only after 1932. During the thirties.

The British Transport Commission conducted a number of studies in the 1950s to go into the aspect of lateral stability of the Long Welded Rails under temperature induced compressive stresses. What this implies is that in long welded panels. free rail expansion or contraction is not being permitted and thereby undergo what are called as 'thermal stresses' due to constrained expansion and contraction.4 Although longer length rails were being used. Studies by the British Transport Commission (Reference 2) however.(2) the weights and lengths of standard rail sections varied from 22kg/m to 65 kg/m and from 5. It was only after the lateral stability of rails under compressive forces was confirmed was the concept of Long Welded Rails accepted by track engineers. the expansion gaps required at the ends of these panels.1. The length of welded panels varied from 18m to 380m. Interestingly it was found that these welded panels derive resistance to buckling not only from the stiffness of the rail itself but also from the rail sleeper fastenings and ballast resistance. The ability of the welded panel to withstand compressive stress buildup due to constrained expansion was another source of worry to the early track engineers. track engineers were unsure of the expansions which these longer length rails would undergo under temperature variations and consequently. . South Africa and elsewhere.5 m to 27m respectively (Reference1). The welding process most commonly used at that time was ‘thermit' though flash butt and other processes were in use in America. 1.indicated that the gap to be provided is not proportional to the length of the welded rail and in fact the expansion gaps are actually independent of the length of the rail.

1.(3) 1. ii) elongated fish-bolt holes. the GIP Railway had undertaken welding of rail joints using the electrical process. other railways viz. The phenomena of battering and hogging of rail joints is shown in Fig. From 1947 to 1966 large number of 5-rail panels (65m in BG and 60m in MG) and 10-rail panels (130m in BG and 120M in MG) were put into track.1 Scale Rail Rail ← Hog → Rail Batter Fig. large scale maintenance problems were reported by various railways regarding the behaviour of 5-rail and 10-rail panels due to: i) increased rail battering and hogging. NW. In 1939. iii) bent fish-bolts.1 . Between 1940 and 1946.1. 1. GIP and EI railways also commenced trials with welded joints. BN Railway started conducting trials on welded rail joints. The purpose was to reduce the maintenance efforts by reducing the number of joints. However.5 In India in the nineteen thirties.

LENGTH : ONE BLOCK SECTION LENGTH LWR Fig.2(b) f CWR CWR g S PECIAL A RRANGEMENT FOR TRA NSFERING FORCE FROM S TOCK RAIL TO TO NGUE RAIL A ND WING RAIL TO CROSS ING CWR CWR Fig. 1. 1.(4) SWR SWR LENGTH ≤ 39m in BG ≤ 36m in MG SWR Fig. 1.2(a) SEJ SEJ B REATHING L ENGTH TOTA L L ENG TH OF LWR SEJ SEJ CENTRAL PO RTION PORTION CENTRAL B REATHING L ENGTH Minimum length of LWR = 250m in BG = 500m in MG MAX.2(c) .

will behave like an SWR.6 The findings and recommendations of the committee make interesting reading (Reference1). 39m in BG and 36m in MG.2(a)) LWR : (LWR) is defined as a welded rail whose central portion does not exhibit any 2) .Rly to investigate into the behaviour of 5-rail panels and 10-rail panels at the first instance and thereafter of Long Welded Rails. temperature and ballast conditions for laying the LWR. recommended that: 1) 2) 3) Welding of 5-rail and 10-rail panels was to be discontinued. therefore. While the capacity of the IRS fishplate design is to accommodate a movement of 15 mm. The 3-rail panel therefore appears to be roughly the longest rail which could be laid in the track with the conventional fish plated joints. Joint Director Standards (Track).(Fig1. 1. Existing 5-rail and 10-rail panels were to be cut into two and half rail panels. The committee found that the IRS fishplate design as per current standards is inadequate to cater to the expansion and contraction occuring in 5-rail and 10rail panels.(5) Taking cognizance of these problems. RDSO was to conduct further studies for deciding the track structure. the Railway Board vide their letter No 65/W6/WRL/6 of 20th January 1966 appointed a committee consisting of 3 Engineers in Chief (Track). The committee. RDSO and a Deputy Chief Engineer of W.1. the actual movements of a 5 rail or 10 rail panel are much larger resulting in large gaps. 1.2 Some Basic Definitions 1) SWR : A rail which expands and contracts throughout its length: 3-rail panels. bent fishbolts and elongated fishbolt holes.

2(c)) Rail Temperature : The temperature of the rail as measured by an approved rail temperature measurement device. Switch Expansion Joint (SEJ) : A physical device placed at the end of the breathing length of an LWR to accomodate the expansion and contraction of the breathing length. 3) 4) 5) 6) 7) 8) 9) .(6) longitudinal movement on account of temperature variations. Stress-free Temperature or Neutral Temperature (tn) : The rail temperature at which an LWR is free of longitudinal thermal stress is called the stress-free or neutral temperature. Destressing Temperature (td): The average rail temperature at the time of fixing the rail fastenings after a destressing operation performed manually is called the destressing temperature.e. The movements are exhibited by track lengths on either side of the central portion which are called as ‘breathing lengths’. (Fig 1. While the maximum length of an LWR is restricted to one block section length. exhibits movements due to temperature variations is called the breathing length. Breathing length : The lengths at the ends of an LWR on either side where the LWR "breathes" i. the minimum length of an LWR should be 500m in MG and 250m in BG.2(b)) CWR : An LWR which continues through station yards including points and crossings is described as a continuous welded rail. Buffer rails : A set of free rails used in lieu of SEJs as a temporary measure to accomodate the movement of the breathing length of the LWR. (Fig 1.

It has been estimated that the magnitude of thermal force in an SWR may be as high as 70% to 75% of the thermal force in an LWR. The SWR therefore develops thermal forces because of being subjected to constrained expansions and contractions. l Fish plated joints are a source of large dynamic forces. This coupled with fish plated joints makes the SWR very vulnerable and should therefore be maintained with care. Sabotage at fish plated joints has been a major worry for the Indian Railways and this has again come into prominence with the accident of the Rajdhani Express at Rafiqganj in Bihar in which a large number of people lost their lives and which was attributed to removal of fishplates on the approach of a bridge. economical and comfortable due to the reasons listed below : l LWR tracks eliminate fishplated joints leading to safety.3 An Explanatory Note on the Short Welded Rail Indian Railways have adopted a 3-rail panel (39m in BG and 36 m in MG) as the Short Welded Rail(SWR) with fish plated joints due to reasons mentioned in 1. l Due to development of large dynamic forces at the rail joints the track geometry at the rail joint gets disturbed . Although the SWR is exhibiting expansion and contraction throughout its length.(7) 1. it is not a free expansion or contraction due to the action of ballast and fastenings.6. As a result fish plated joints exhibit large scale rail wear and development of cracks from fishbolt holes and fractures. 1. In some instances premature rail renewal may have to be carried out due to excessive fractures. The reason for the widespread popularity of the LWR are the numerous advantages which are derived using a LWR track structure. The provisions of laying and maintenance of SWRs is given in para 505 of the Indian Railways P-way Manual (1986).4 Advantages of the Long Welded Rail Today the LWR is synonymous with modern track.1. The LWR makes train travel more safe.

l Due to impact at rail joints there is an added wear and tear of rolling stock wheels to an extent of 5% and as the wheel has to negotiate the gap there is added fuel consumption to an extent of 7%.(8) frequently resulting in an increment in the track maintenance effort. It has been estimated that there is as much as 25% to 33% savings in the track repair and maintenance costs due to elimination of rail joints. l Due to elimination of noise and vibrations at the rail joints passenger comfort is substantially increased. .

it will undergo a change in length equal to L α t.1 Basic Principles A metal rod supported on frictionless rollers can theoretically expand and contract freely with variations in temperature. where α is the coefficient of linear expansion and t is the change in temperature. Creep resistance further offered by the railsleeper fastening.(9) CHAPTER II PRINCIPLES OF LONG WELDED RAIL 2. if there is a steel rod 13 m long (the same as the standard length of a rail). Creep resistance on account of friction between the rail and sleeper at the rail seat. Thus with any change in temperature it is not the rail alone but the rail-sleeper frame as a whole which tends to move.152 x 10-5 per degree centigrade) However. For example. 2. Thus the expansion or contraction of the rail is less than what it would undergo if it was completely unrestrained. If the length of the metal rod is ‘L’.152x10-5)x20 = 3 mm. . thereby preventing any relative movement between rail and sleeper. then it will undergo change in length equal to 3 mm. if the temperature of the rod changes by 200C as can be seen from the calculation below: Expansion/Contraction=Lα t = (13x1000mm)x(1. 1. (α =1. the rail in the track cannot be compared to the metal rod supported on frictionless rollers as mentioned in the preceding paragraph. In LWR the rail is held down to the sleeper by fastenings which have adequate toe load. The rail is restrained from free expansion and contraction by the sleepers because of.

b) The temperature at which the track could be attended under regular track maintenance. are governed by the sole consideration that the thermal forces at any time in the track should be within safe limits. If the temperature difference becomes more. The frame is under restraint because of the resistance offered by the ballast in which the sleepers are embedded. then the rail would expand or contract with variation in temperature and consequently no force would build up in the rail. If the track frame was not restrained. . Longitudinal ballast resistance is assumed to be uniform for sleepers of the same type. etc. since there is a restraint now offered by the longitudinal ballast resistance. then a small length of track at the end would be sufficient to develop longitudinal ballast resistance against the tendency for free movement of the rail. However. These are called thermal forces. This limit is dictated by: a) either the maximum or the minimum rail temperature and the temperature at which the rail was fastened to the sleeper. a limit up to which the temperature differences can build up. If the temperature variation (from the temperature at which the rail was fastenced to the sleeper) is small. There is however. Consequently. a longer length of track at the ends would be called upon to develop the necessary longitudinal ballast resistance against the free movement of the ends of the rails and there would be a corresponding increase in the thermal forces. However. to avoid the eventualities of buckle and fracture of the LWR. forces are set up in the rail. the temperature at which the patrolling of the track should be introduced. it would lead to a small amount of thermal force in the rails.(10) Here again the rail sleeper frame is not entirely left unrestrained. The resistance offered by the ballast to the movement of the track frame in the direction of the track is called longitudinal ballast resistance. This longitudinal ballast resistance builds up progressively from the ends of the long welded rail towards the centre.

which is the temperature at which the LWR is free of longitudinal thermal stress. If the central portion of length ‘L' had been free to expand it would have expanded by an amount equal to ‘Lαt’. Let us assume the LWR central portion to undergo a change of temperature equal t oC temperature rise. or If P is the force induced in central portion (compressive force) and A is the cross section area of rail. and E is in N/ mm2. A is in mm2. a technically correct formula for the thermal force in the LWR will be P = A E α (tP-tn) where tP is the prevailing rail temperature and tn is the rail neutral temperature. at which the LWR was laid destressed and α is the coefficient of linear expansion. An expression for the magnitude of this thermal force is required as this will govern the design of the LWR. However since the central portion of the LWR does not move. As t represents the change of temperature of the rail w. Since stress /strain = E (Young’s modulus of rail steel) we have P/ A =E αt or P = A E α t where P is in Newtons.r. the temperature it is stress free.2 Force Diagram In the previous paragraph it has been indicated that thermal forces are built up in the LWR due to the resistance offered by the ballast to the free expansion or contraction of the track frame. the compressive strain induced in central portion is equal to L αt = αt where ‘t’ is the change of temperature of the LWR L with respect to the temp. . then P/A is the compressive stress in the rail.t.(11) 2.

(12) When tP = tn.5 MPa/0C For 300 C change of temperature therefore thermal stress induced will be about 75 MPa which is about 8. it can be seen that force developed in the LWR depends primarily on the prevailing rail temperature and the rail neutral temperature. 2.5% of the ultimate tensile strength of a 90 UTS rail.3 Importance of Rail Temperature From the expression for force in an LWR.1 Thermal stress in rail due to unit degree change of temperature = E x α = 2. The force at the beginning of the LWR is zero. 2.15 x 106 X 1. The shape of the force diagram is therefore as given in Figure 2. P is a tensile force. As tn is not known it is assumed that tn = tL or td .1 Fig.152X10-5 = 2. temperatures at which LWR was laid (tL)or destressed (td).tn). This change of force from zero to a peak value occurs over the breathing length. P = AEα(tP . when tP>tn. P=0. Both these factors are discussed below: . P is a compressive force and when tP < tn. and in the central portion equal to A E α t.

Where a number of thermometers are used to measure the rail temperature at one place.(13) 2. appreciably different from the other adjoining thermometers. as in case of laying of LWR. destressing etc. shall be considered as defective.3.1 Rail Temperature Measurement : Thermometers The following are the different types of approved thermometers for measuring rail temperature : i) Embedded type .It consists of a graduated chart mounted on a disc which gets rotated by a winding mechanism at a constant speed to complete one revolution in 24 hours or 7 days as applicable. A steady recording of the rail temperature is reached within 8 minutes. giving a continuous record of rail temperature. Continuous recording type . Dial type .This is a bi-metallic type thermometer which is provided with a magnet for attaching it to the rail. ii) iii) iv) Any other type of thermometer approved by RDSO/ Chief Engineer. This type of thermometer takes 25 to 30 minutes for attaining temperature of the rail. the cavity filled with mercury and sealed.This is an ordinary themometer inserted in a cavity formed in a piece of rail-head. . The sensing element is attached to the web of the rail and connected to the recording pen. The rail piece is mounted on a wooden board which is placed on the cess and exposed to the same conditions as the rail inside the track. any of the thermometers showing erratic readings. The thermometer is attached on the shady side of the rail web as this location is approximating the average rail temperature to the greatest extent. through a capillary tube wihch is filled with mercury.

and the corresponding air temperatures obtained from the Meteorological office. it was possible to determine the maximum and minimum rail temperatures obtainable at these stations. RDSO conducted rail temperature studies between 1969 and 1971 over a two year period. On these stations rail temperature records shall be built up using preferably a well calibrated continuous recording type thermometer. (Fig. 22 stations were identified over the Indian Railways where Standard Measuring Arrangements for Rail Temp (SMART) were set up. 2. The rail temperature was measured by means of a thermometer placed in a mercury-filled hole in the rail head.(14) Zonal Railways should nominate 8 to 10 stations in their railway in a manner as to give the representative sample of the temperature variations on the Zonal Railway for the region allotted to each station. This data was presented in the form of a rail . Reference-3) for each of the identified 22 stations. Using these correlation equations and the maximum and minimum air temperatures at 180 stations over the Indian Railways. Correlation equations between the rail temp and air temp were derived using a computer based regression analysis (Details available in RDSO/C/ 146. 2. obtained from the Weather office over a period of 90 years. Rail thermometers shall also be available with each each gang and sectional PWIs to enable the gangs to work within the prescribed temperature ranges.2 Rail Temperature Zones and RDSO Studies : In order to understand the correlation between the rail temperature and ambient temperature. These stations shall be the existing PWIs offices. destressing operations and requirements of hot/ cold weather patrolling etc. The maximum and minumum rail temperature for a continuous period of at least 5 years shall be ascertained and the mean rail temperature (tm) for the region arrived at.2) SMART consisted of a full-length rail laid in the east-west direction on wooden sleepers with ACB plates and boxed with standard ballast profile.3. These temperature records shall be analysed to assess the probable availability of time periods during different seasons of the year for track maintenance. Rail temperature readings were taken on an hourly basis between August 1969 and August 1971.

(15) Fig. 2.2 : STANDARD MEASURING ARRANGEMENT FOR RAIL TEMPERATURE. .

namely the rail temperature range and the rail temperature mean.(16) temperature map where against each station two figures are given. As the rail temperature rises compressive thermal forces will be built up and when the rail temperature reaches tmax compressive forces proportional to the full range of rail temperature will be built up. Refer to Fig 2.3. The Rail Temperature Range = Max Rail Temp . The block shows the maximum. .Min Rail Temp and Mean Rail Temperature = Max Rail Temp + Min Rail Temp 2 The country is further divided into four zones depending upon the temp range : Temp.3 Rationale behind choice of tn or td The LWR neutral temperature should be chosen in such a manner that the thermal force developed in the LWR is within the desired limits. Zone I II III IV Temp. Let us see what happens if we fasten the rail to the sleeper at the minimum rail temperature(tmin). In this case there is of course no danger of any tensile force developing in the rail and consequently of rail fracture.3. minimum and range of rail temperature in the four temperature zones. at a temperature anywhere within the range between maximum and minimum rail temperatures. Range 40-50oC 51-60oC 61-70oC 71-76oC The rail temperature map is given in the LWR Manual (1996). 2. Such large compressive forces could be very dangerous to the stability of the LWR and the track can buckle. The rail can be fixed to the sleepers by fastenings after destressing.

.3 : VARIATION OF TEMPERATURE WITH RESPECT TO td IN DIFFERENT ZONES.(17) tmin tmin tmin tmin Fig. 2.

Since 90R and lighter sections do not have adequate margin to accommodate the resulting thermal tensile stresses. td for such rails has been fixed between tm and tm + 50 C. since the rail temperature can fall through the complete range of temperature at this place. if the track buckles due to excessive compressive forces in the rail. Logic suggests that we should fix the destressing temperature exactly mid-way between maximum and minimum rail temperatures. A fracture will create a gap in the rail. However. All the same compressive forces and tensile forces are bound to develop and so the possiility of buckling as well as fracture will exist. at mean rail temperature.e. other conditions remaining the same. and consequently there is no danger of buckling. However. However. . in the case of fractures. This is the basis for fixing td on the lndian Railways at a temperature above tm. tensile stresses and forces in the rail could develop to a very large magnitude making rail fracture very probable. Thus at least a few trains can pass over a fractured rail without accident till the fracture is attended to. i. Heavier 52 kg and 60 kg rails having greater section modulus can accommodate relatively larger thermal tensile stresses and so td for these rails has been fixed between tm + 5 and tm + 100C. the alignment of the rail is not immediately distorted. In that case the extent of maximum compressive or maximum tensile forces would be equal and half of what it would otherwise be as in case of either of the two previous situations. It also gives relief by reducing the compressive forces which are directly proportional to area of cross section of the rail. Also fractures rarely occur on both rails and at the same location simultaneously. though at the cost of introducing higher tensile forces. Since temperature cannot rise any further. alignment of track gets distorted and safe running of trains is endangered. there is no likelihood of compressive thermal stresses developing in the rail. Therefore considering buckling more dangerous. it is considered prudent to fix the destressing temperature higher than the mean rail temperature so that the compressive forces built up in the track would be within reasonable limits.(18) Let us see what happens if the rail is fastened to the sleeper at the maximum rail temperature (tmax).

If td is fixed at a lower level. Again if td is set at a lower temp. This would either entail limited working hours to the gangs. night working or working in split shifts. The LWR manual prescribes imposition of speed restrictions at locations where track maintenance activities have been carried out and the temperature rises beyond td + 200 C during the period of consolidation. a range has been recommended for td instead of a fixed value.(19) Since the operation of fastening of the rails to the sleepers after destressing takes time during which rail temperature can vary. Hence the clause of the LWR Manual limiting the maintenance operations on the LWR to a rail temp of td + 100C has been provided. The other provision of the manual is that hot weather patrolling should be introduced when tP rises above td + 200C. the prevailing rail temperature could rise above td + 200 C quite often leading to an increase in the quantum of hot weather patrolling. restricting the maintenance operations. If the destressing temperature of the rail is lowered the prevailing rail temperature could rise beyond td +100 C more frequently reducing the availability of maintenance hours. In addition to the above consideration of avoiding buckling while risking the chances of fracture there are some more reasons why the destressing temperature has been fixed at a level higher than tm. 1) Studies on LWR behaviour have indicated that the stability of the track gets endangered at temperatures above td + 100 C. and ensuring adequate maintenance hours. These options have serious practicability problems. as compared to tm. Hence the destressing temperature of the LWR has been fixed at a temperature above the mean temp so that the prevailing rail temp does not frequently go above td + 100 C. this would necessitate imposition of a large number of 2) 3) .

It is denoted by R and its unit of measurement is in kg/metre/rail.(20) restrictions at work sites. 2. then LB x R represents the total resistance offered by the ballast in the breathing length. whenever the rail temperature rises beyond td +200C during the period of consolidation. Keeping the above factors in mind td has been fixed at a higher level as compared to tm.4 Breathing Length 2.1 As discussed earlier the force build up in the LWR starts in the breathing length at the free end.4. The longitudinal ballast resistance is mobilized when there is a relative movement between the sleeper and the ballast. A more realistic view would be to have a flexible neutral temperature which could be shifted downwards in winter to control fractures and shifted upwards in summer to control the tendency of the track to buckle. The tendency of the rail sleeper assembly to expand or contract is resisted by the resistance of the ballast called the longitudinal ballast resistance. Fractures are today one of our major concerns and there is a demand from some railways to reduce the level at which the destressing temperature should be fixed basically to reduce the number of fractures. As the maximum force in the LWR = AE α t LB x R = AE α t or LB = AEαt R The above expression for the breathing length indicates the various factors which govern the breathing length as under:1) The breathing length is proportional to the temperature . If LB is breathing length and R the longitudinal ballast resistance.The policy makers at that time knew that doing this will have an effect on the number of fractures and therefore indicated that a greater frequency of USFD inspection and monitoring would be required. This situation is unacceptable.

These studies are described in RDSO Report No. The load applied was measured with the help of a proving ring and the longitudinal movement of the panel was recorded with the help of dial gauges. The load was applied to this test panel by a hydraulic jack with a remote controlled pumping unit. 2) 3) Annexures 1A. After the test.(21) change. C-148(Reference 4). Experimental Set up : These studies were conducted experimentally on a running track as well as on a freshly laid track in the lab. (Refer to figure 2. The larger the cross sectional area of the rail. The larger the value of 'R' the longitudinal ballast resistance.5 Longitndinal Ballast Resistance The longitudinal ballast resistance ‘R’ comes into play when there is relative motion of the sleepers with respect to the ballast in the longitudinal direction. The instantaneous loads and movements were measured as the load was increased gradually till it reached a peak and fairly steady value.5. As BG sleepers have a larger value of R the breathing length of BG LWRs is smaller as compared to MG LWRs. 1B & 1C of LWR Manual give indicative values of breathing lengths for different sleepers and sleeper densities in different zones. the smaller the breathing length. 2.1 RDSO conducted a number of experimental studies on the various aspects of the longitudinal ballast resistance. the short length rails were replaced by the normal rails. The tests were conducted on a running line where a traffic block of 90 minutes was taken. Therefore the breathing length is maximum in Zone IV and minimum in Zone I.4) The maximum value obtained from the proving ring was divided by twice the number of sleepers in action to get the ballast resistance . 2. A track section comprising of short length rails and three sleepers embedded in ballast was pushed in the longitudinal direction and the displacement versus load curve plotted. the larger the breathing length.

4 .(22) Fig 2.

to 1500 nos. A different . This increase is maximum for concrete sleepers. A heaped up shoulder ballast gives a higher ballast resistance as compared to the standard shoulder for both BG and MG. The effect of traffic on the growth of the ballast resistance is substantial. Ballast resistance per sleeper decreases as the sleeper spacing reduces. 5. In BG. the order is PRC. 4. CST-9 and Wooden.(23) per sleeper per rail. 3. Through packing causes a reduction in ballast resistance. Findings of the study : 1. per km. For larger sleeper densities the value of the longitudinal ballast resistance again increases due to heavier track structure. 6. This figure divided by the sleeper spacing in metres gives the value of 'R' in kg/m/rail. In consolidated and through packed conditions other sleepers in order of decreasing longitudinal resistance are: Steel. RCC twin block. Table2 and Table3. For BG conventional sleepers the reduction is 36%. Wooden and CST-9. For both BG & MG concrete sleepers the average reduction is 23%. RCC. summary of values obtained for different sleepers under conditions is given in Table1. 2. BG concrete sleepers attain 86% of the consolidated value on passage of 1 GMT of traffic. Steel. PRC sleepers give the highest longitudinal ballast resistance in all conditions. but the ballast resistance per unit length of track remains more or less constant for sleeper densities from 1200 nos. Deep screening causes a greater reduction in ballast resistance. In deep screened or freshly laid track.

(24) Table 1 Longitudinal Ballast Resistance (kg/m/rail) (Effect of sleeper type and track maintenance activities) Gauge Sleeper Type BG PRC Steel CST-9 RCC Consolidated 1244 1051 933 921 Through % packed Loss 1027 744 551 666 17 29 41 28 Deep % Screened Loss 885 433 276 581 29 59 71 37 47 36 30 Wooden 697 MG Wooden 405 Steel 298 380*/552 45*/21 370 180*/302 55*/21 259 231 22 209 * Values are for tracks packed by MSP Conventional sleepers are beater packed PRC sleepers have been off track tamped. Sleeper spacing PRC – 1600 Nos/km RCC and conventional – 1200 Nos/km Adapted from RDSO/C-148 .

9 RCC 1244 1051 933 921 .(25) Table 2 Maximum Longitudinal Ballast Resistance (kg/m/rail) in freshly laid conditions (Effect of sleeper density & heaped up ballast profile) 38 mm Standard Ballast profile A B C D 38 mm Heaped up Ballast profile A — — B — — C 648 438 D — — Ballast Sleeper PRC Steel 546 559 561 628 347 375 387 434 A : 1200 Nos per km B : 1400 Nos per km C : 1600 Nos per km D : 1800 Nos per km Adapted from RDSO/C-148 Table 3 Longitudinal Ballast Resistance (kg/m/rail) (Effect of movement of traffic) Sleeper consolidated Value Resistance on passage of 1 GMT of traffic 1072 740 580 750 Adapted from RDSO/C-148 % of consolidated Value 86 70 62 81 PRC Steel CST.

6 Lateral Ballast Resistance The lateral ballast resistance comes into play when the track has a tendency to get displaced in the lateral direction due to build up of compressive forces. The test setup is given in Fig.5 to 1 inch) can cause significant reduction in lateral track strength.(26) 2. RDSO studies(Reference 5) conducted on various aspects of lateral ballast resistance have indicated the values of lateral ballast resistance as given in the Table below.5. 2. 2. The reduction in the lateral ballast resistance on through packing and deep screening is quite significant for CST-9 and steel sleepers and much less for PRC sleepers. Tests conducted have revealed that : u Track surfacing and ballast tamping even with a minor amount of rail lift (0. The higher resistance recorded by metal sleepers is due to the central keel in the case of CST-9 sleepers and turned down ends in case of steel sleepers which get embedded in the ballast core and offer better resistance to the lateral movement. Values of Lateral Ballast Resistance in Kg/m of track Gauge : BG Sleeper Consolidated Through Packed 1100 1226 825 1040 320/520 Deep Screened 532 1040 540 800 384 CST-9 1640 PRC 1470 Steel 1430 RCC 1420 Wooden 1060 Adapted from RDSO/C-156 1. .

3 GMT. For instance. 2. Dial gauge f Fig.5 . the dynamic track stabilizer may produce a consolidation equivalent to 0. Dynamic track stabilizers could significantly acclerate ballast consolidation or strength recovery. for granite ballast.(27) u u Depending upon the ballast type. recovery of strength loss due to traffic could vary from 0.3 GMT to 9 GMT.

3. P(x) Fig.(Fig. It is possible to make certain simplifying assumptions and estimate the movement at any point in the breathing length. At the ends of the LWR since the restraint offered by the longitudinal ballast resistance is nil.(28) CHAPTER III THERMAL MOVEMENTS AND HYSTERESIS 3.3.1) At B. As the longitudinal ballast resistance exerted on the sleepers progressively builds up complimentary forces in the rail increase from A towards B. The movements recorded in the field at various points in the breathing length of the LWR corroborate the above mentioned observations.1 . the movement reduces to zero. the movement is observed to be the maximum.1 Estimation of Thermal Movements It is only in the breathing lengths that a LWR displays the property of longitudinal movements. which is the junction between the breathing and fixed lengths.

If this net expansion is called dy. The net expansion of the small length of rail dx will therefore be the difference between the above two values. is equal to P(x) dx EA (where P(x) is the thermal force present in the small length of rail dx at a distance x away from B). Thus the amount of maximum contraction or expansion at any point .(29) Take a small length of rail dx at any arbitrary point M at a distance ‘x’ away from B (refer fig. The amount of contraction of this length dx. then dy = αtdx − P ( x ) dx AE = AEαtdx − P( x)dx  P − P( x)  = dx AE  AE   where P= AE α t is the maximum thermal force in the central portion of the LWR. 3. Integrating the net expansion of all such small lengths of rails starting from B towards M. we can obtain the total expansion or displacement of points M as Y = ∫ dy = 0 x 1 AE ( ∫ [P − Px)]dx x 0 It can be observed from Fig. 3. It is possible to calculate the amount of free expansion of the small rail of length ‘dx’ due to change of rail temperature as well as the amount of contraction due to presence of thermal force present in the rail at that length: Free expansion of this small length dx due to a rise in rail temperature by t0C = dxα t.1).1 that the expression (P-P(x))dx is nothing but the area of the shaded diagram appearing above the diagram of thermal force.

(Equation 2) We can look at it this way : The maximum movement of the end of the LWR is half the corresponding value if only the breathing length LB of the LWR is allowed to expand or contract absolutely freely.3. m = ((LB) α t)/2 . upto that point as in the Fig. at ‘A’ or ‘D’ can be obtained as follows: Maximum expansion or contraction at ‘A’ or ‘D’ = AreaofTriangleA1 FE AE = 1/2 x P(LB)/AE . m = (LB α t )/2 can also be rewritten as: ... LBαt 2 l l l l Breathing f f Length Central portion Variation of movement in Breathing length Fig..2..1 by EA. equation (1) above can be simplified as maximum movement at the end of a LWR.(30) in the breathing length of a LWR can be computed by dividing the shaded area from B. (Equation 1) As P= AE α t.e. Extending this logic.2 This expression for the maximum movement at the end of the LWR i.e.. 3. The variation of movement along the breathing length is given in Fig. the cumulative value of expansion or contraction at the end of the LWR i. 3..

28 kg/cm/rail A = 66. (Equation 3) This can be derived from Equation 1 by substituting AEα t/R in place of LB because LB = P/R = AE α t/R It should be noted that in the above calculations an important assumption has been made that in a breathing length of LWR. Zone IV with temp.15 cm2. the sleepers have equal values of longitudinal resistance. Rails: 52kg (A=66. fall α= 1..15 × 106 ( .(31) m = ΑΕ (α t)2 /2R .57 mm (forward movement of tip of tongue/stock rail wrt reference line) .28 = 5. Sleepers: PRC.15 x 106 kg/cm2 t = t + 10 C d m 0 t = 280C for temp. ∆= AE (αt ) 2 2R 2 66.. E = 2.152 × 10−5 × 28) 1 ∆1 = 2 × 13. rise t = 480C for temp.15cm2) Sleeper density 1540 sleepers/km. To illustrate the above let us solve an example with the data given below: Gauge: BG.152 x 10-5 /0C As discussed earlier movement at an SEJ For a temperature rise of 280C. range =760C R = 13.15 × 2.

88 mm The movements which occur at the SEJ due to thermal variations are shown in Fig.15 × 2.3 (A. . 3.37) = 2 x 21.152 × 10 −5 × 48) 1 ∆2 = 2 × 13.15 × 106 ( .3 For a temperature fall of 480C 66.(32) At tp=td At tp=td+280C At tp=td-480C Fig. 3.37 mm (backward movement of tip of tongue/stock rail wrt reference line) Total movement at the SEJ joint = 2 x (5.28 2 = 16.94 = 43.B.C) with values as calculated in the above example.57 + 16.

with 6 special sleepers to RDSO drawing No. RDSO/T-4160 Description Assembly for Switch Expansion Joint with 80 mm max gap for LWR BG 52 kg on Concrete sleepers Assembly for SEJ with 80 mm max gap for LWR BG 60 kg on concrete sleepers Assembly for SEJ with LWR BG 60 kg (UIC) on PSC sleepers laid on curve with curvature from 0. An SEJ typically consists of a pair of tongue rails and stock rails .50 Assembly for SEJ with 190 mm max gap for bridge approaches for LWR BG 52 kg on concrete sleepers Assembly for SEJ with 190mm max gap for bridge bridge approaches for BG 60 kg(UIC) on PSC sleepers Assembly for SEJ for 80 mm max gap with BG CR – 120 crane rails on PSC sleepers RDSO/T-4165 RDSO/T-5748 RDSO/T-6039 RDSO/T-6263 RDSO/T-6257 (i) RDSO/T.50 to 1. All these are 300 mm wide sleepers with sleepers Nos 10 and 11 . the tongue rail laid facing the direction of traffic. The distance between the tip of the tongue rail and notch of the stock rail is typically kept as 40mm at the destressing temperature. The tongue rails and stock rails are machined and given suitable bends to accomodate each other. Earlier the gap used to be 60mm and has now been reduced to 40mm to lower the impact of the passing wheel.50 . Modern SEJs are laid on concrete sleepers with rail free fastenings.(33) 3.4165 are the conventional straight SEJs with 80 mm maximum gap. Tongue rails and stock rails have typically straight alignment and hence these SEJs cannot be laid in curves sharper than 0. Each SEJ has a pair of tongue rails and stock rails.RDSO/T4149. Various types of SEJs used on Indian Railways are described below: Drawing No.4160 and RDSO/T.2 Switch Expansion Joints (SEJ) : The thermal movement in the breathing length of an LWR are accomodated at the switch expansion joint (SEJ).

4 A l l Angle ties (34) . 3.Typical SEJ layout Fig.

3.4 (a) .RAIL REFERENCE POST (35) Fig.

10 and 11 with special fastenings. (ii) RDSO/T-6039 & RDSO/T-6263 : These are wide gap SEJs for bridge approaches where the maximum gap permitted is 190 mm. 8.5 mean position is kept at 166 mm from centerline of sleeper No. Sleepers other than 7 to 13 are approach concrete sleepers with normal fittings. . Fig.4(a) gives the details of location A. 12 and 13 with similar fittings. 10 coincides with the tip of the tongue rail and the 40 mm initial gap is provided with the tip of the tongue rail coinciding with the centre of the sleeper No. 10 and 11 is 700 mm while the sleepers spacing from 1 to 10 and 11 to 20 may be 600 mm or 650 mm depending upon the sleeper density. 8. use of an insertion piece in the gap should be made. 10.5o & < 1o p p 11690 mm curvature > 1o & < 1. 9. 11. 3. Use is made of ERC Mark II clips with flat toe designed with a toe load of 350400 kg/clip to enable free rail movement. 13 are special sleepers with sleeper nos. 10. 12. The centre line of sleeper No.4 gives details of a typical SEJ layout and 3. the curvature > 0. Sleepers No. 10 to enable the tongue rail to remain on the sleeper even when the entire expansion takes place. 7.(36) with special fastenings and sleeper No. The centre to centre spacing of sleeper No. 9.5o p p 11690 mm Fig 3. When gap is more than 100mm for passing diplorries with smaller wheel diameter. 9.

3. The non-bolted segment of the gap avoiding rail braces the machined segment of the tongue rail. A brief description of these layouts is given below: (1) SEJ with one gap : This design has been developed by M/s Rahee Industries Ltd. Calcutta.6 ONE GAP SEJ .50 but not sharper than 1. Improved design SEJs developed by various industries are under trial on the Railways till final approval is received from the Railway Board. This rail is securely fitted to one of running rails with high tensile steel bolts.(37) (iii) RDSO/T-5748 : These SEJ layouts can be used when the SEJ has to be laid in a curve sharper than 0. Fig 3. The other running rail is called the tongue rail. The conventional SEJ design involves two bends in the stock rail and tongue rail which are locations of weakness resulting in fractures. The tongue rail and stock rail are given curvature as given (Fig. This running rail together with the gap avoiding rail is called the stock rail.6 The design comprises of a pair of machined segments on the non-gauge face side of two non-bent running rails mounted with a gap between the juxtaposed rail ends and a third rail called a gap avoiding rail of predetermined length accommodated in the said machined segments parallel to and adjacent to the non bent straight length of the running rails.5).50. Gauge Face Side → HTS BOLTS 11 1 Fig 3.

Salient features of Bina Metalway 2-gap SEJ (Fig 3. with three sleepers located near each gap. 1 to 31 should be at a spacing of 600 mm c/c. 11. T/4149. In both these designs two gaps of maximum 80 mm each are provided in one SEJ. 12 and 22 with the tip of the tongue rail coinciding with the centre line of the sleeper. (2) SEJ with two gaps Two designs have been developed by two different firms (M/s Bina Metalway. Two cut rails are joined together to make the stock rail.7) The stock rail is considered to be static with negligible expansion and contraction in length due to temperature changes. 23 and 24 are special sleepers to RDSO drawing No. Similarly a gap of 80mm is available for the LWR on the other side. Hence a total gap of 18750 mm should be created while inserting this SEJ. 12. Sleeper Nos. T – 4149 and the rest are normal PSC line sleepers. 1. Check rails guard against excessive play of worn out wheels. 22. . 10. Design suitable upto 180 mm maximum gap. The length of the SEJ is 5750 + 6950 + 5920 + 80 = 18700 mm.(38) Advantages : 1) 2) 3) 4) No bends in tongue and stock rail. While laying the SEJ it should be ensured that the ends of the stock rail are 40 mm away from the centre line of sleeper Nos. Thus a maximum gap of 80 mm is available for an LWR on one side of the SEJ. The tongue rail is manufactured by cutting the rail at head and foot location. Sleeper Nos. The stock rail is fabricated out of two pieces of lengths 7140 mm and 5920 mm connected to each other by HTS bolts. Jamshedpur and M/s Chintpurni Engineering Works. This SEJ makes use of 6 wider concrete sleepers each to Drg No. 2.Barabanki). Only 5 special sleepers of standard SEJ on PSC assembly are used.

and stock rail end kept at 40 mm from mean position. thus creating a gap of 40 mm. 5. as far as movement at the SEJ is concerned. 12 and 22. 3. 6. 12 and 22.3 Phenomenon of Hysteresis : The behaviour of an LWR. The mean gap is 40 mm on each end. Mean position of SEJ should be kept at centre line of sleepers No. rise curve OF where ) ) ) ) . The tongue rails are kept at mean position at centre line of sleeper Nos.7 TWO GAP SEJ 3. uniformly rises above ‘O’. the movement or expansion at the SEJ follows the movement – temp. The mean position should also be marked on the rail posts erected on both sides of track.(39) Tongue rail ) ) stock rail ) ) Tongue rail FIG 3. is irrational which will be evident form the (Fig. 3. 4.8) shown below : As the temp.

(40) F D E B C A1 Fig 3.8 HYSTERESIS PHENOMENA .

Implications of hysteresis: At temp. This would mean that the gap at the SEJ could be (20 – a) or( 20 + b) if 20 mm is the initial gap. for different zones as given on the adjoining page : . In order to simplify matters. tP the movement at the SEJ may be an expansion equal to ‘a’ or a contraction ‘b’. max min While plotting these curves it should be remembered that whenever there is a reversal of temperature. The final hystersis loop shall be an envelope of the two hysteresis loops as drawn above. 2) Temp fall from td to t . then the movement at the SEJ does not follow the original path but traces out a new curve A2DB2. If at B2 the temp again reverses then the path traced out is B2EA 2 rather than B2DA2 Loops in the form A2DB2EA2 are called hysteresis loops and are formed whenever there is a temperature reversal. LWR Manual Annexure V gives the permissible range of gaps at the SEJ for different track structures at different rail temp. due to hysteresis the gap at SEJ is not a discrete value but a range. fall from t to t and again a rise to max max min t from t max min. rise from t to t and then a fall from min min max t to t . (Fig. an annual hysteresis loop or curve is plotted which will envelope all the hysteresis loops formed on a daily basis.(41) AE (αt ) ∆= 2R 2 At any given temperature t4 if the temp starts falling. Hence. 3. the longitudinal ballast resistance should be taken as twice its normal value ( 2 R instead of R).9) This hysteresis loop can be traced in two ways : 1) Temp rise from td to t .

R.9 .(42) ANNUAL HYSTERESIS LOOP SHOWING THE MOVEMENTS OF ENDS OF L. 3. PROVIDED WITH SWITCH EXPANSION JOINTS Fig.W.

(43) SAMPLE TABLE FROM ANNEXURE V OF LWR MANUAL GAPS BETWEEN REFERENCE MARK AND TONGUE RAIL TIP/ STOCK RAIL CORNER OF SEJ FOR VARIOUS TEMPERATURES IN mm FOR BG. PRC SLEEPER.11 Temperature td + 28 td + 25 td+ 20 td + 15 td + 10 td + 5 td td – 5 td – 10 td – 15 td – 20 td – 25 td – 30 td – 35 td – 40 td – 45 td – 48 Zone I – – – 18 18 to 20 19 to 22 19 to 23 20 to 24 21 to 25 22 to 26 23 to 26 24 to 27 26 to 27 27 ---Zone II – – 17 17 to 19 18 to 21 18 to 23 19 to 24 19 to 26 20 to 27 22 to 28 23 to 29 24 to 29 26 to 30 28 to 30 30 --Zone III – 16 16 to 18 16 to 20 16 to 22 17 to 24 18 to 26 19 to 27 20 to 29 21 to 30 23 to 31 24 to 32 26 to 32 28 to 33 31 to 33 33 -Zone IV 14 14 – 16 15 – 19 15 – 21 16 – 23 16 – 25 17 – 27 18 – 28 20 – 30 21 – 31 23 – 32 24 – 33 26 – 34 29 – 34 31 – 35 33 – 35 35 . 52 Kg. VALUE OF R (BALLAST RESISTANCE) ASSUMED = 13.28 Kg/cm/rail AND t d AS PER PARA 1. 1540 Nos/km.

When gaps have been initially laid at 60 mm. The ballast resistance first increases linearly as the sleeper displacement.(44) ‘Y’ ↑ ↑ +R → ‘X’ G -R F Fig. the path traced out at the end of LWR follows an irregular path leading to the hysteresis phenomenon.10. Cause of Hysteresis : Hysteresis is due to the behaviour of the longitudinal ballast resistance. 10 mm should be added to each of these values. Due to this effect. A tolerance of + 10mm is prescribed beyond the given gaps. A plot of the resistance offered by the ballast vis-à-vis the sleeper displacement is as given in Fig 3. then goes into the plastic zone and finally assumes a constant value R. 3. .10 The gaps are with an initial setting of 40 mm. If at this stage the temperature reverses then the value of the longitudinal ballast resistance drops to zero and then becomes (-R) as shown above. This shows that at the time of reversal of temp the ballast resistance mobilized is 2R. The gaps given in the above table are the theoretical gaps.

3. 3. (Fig.12. This gap between the tongue rail and stock rail will be equal to 40mm for 52kg and 60kg rail sections. GAP MEASUREMENT AT AN SEJ Fig.(45) 3. and for other rail sections 60mm as shown. different zones and different prevailing temperatures. A sample page for filling up the movements observed at an SEJ as per annexure XIII A of the LWR Manual is shown in Fig.4 Gap measurements at an SEJ : At the SEJ a reference line is established between the tongue rail and stock rail.11 .11 ) Gaps g1 and g2 are not discrete values but the permissible range has been defined in the LWR Manual Annexure V for different track structures. 3.

11 Observ Permis ed sible Range SEJs at the end of this LWR : SEJ NO.10 Obser Permis ved sible (a) Range Distance (mm) between tongue/ stock rail & Reference line at SEJ No.Annexure XIII A Chart of movement of LWR/CWR No.10.12.10 c/c space Measu Rectifi Remarks between red by cation sleepers the two carried central sleepers out at SEJ No.Sample of SEJ gap measurments. Also refer Fig 3.10 at km—. SEJ No.11 at km— Date of Time of Rail Right or Measure Measure Temp Left Rail ment ment (46) 6. 10 at SEJ No.11 on by date Distance (mm) between tongue/ stock rail & Reference line at SEJ No.11 .2004 10:00 300C Left rail g2 Right rail g4 6.2004 14:00 400C Left rail Right rail g5 g7 Fig 3.10.

1. In goods running lines. 3.(47) CHAPTER IV PERMITTED LOCATIONS AND TRACK STRUCTURE 4. in each specific case.1 External Equilibrium of Curved LWR Track . 2. 4. rail joints may be welded to form LWR if the condition of all components of the track is generally sound and without any deficiency subject to such relaxation as may be approved by Chief Engineer. goods yards. General Considerations : 1. gauge conversions. reception yards marshalling and classification yards. doublings and permanent diversions will be opened with LWR wherever permissible by the provisions of the LWR manual. All complete track renewals (primary) shall provide for LWR/CWR wherever permissible. Existing rails on permitted locations shall be converted into LWR/CWR provided they meet the requirements laid down in the Manual for Welding of rail joints by Alumino thermit (SKV) process. It is a policy that all new constructions. P/R kg/m P R = CURVE RADIUS IN m P = AXIAL COMPRESSIVE FORCE IN LWR IN kg = 2 AEαt R R P Fig.

Fig. On Broad Gauge a shouder ballast width of 500mm has been adopted.2. In order to ensure that the stability of the LWR in curve is the same as in straight the lateral ballast resistance in curve should be made high by at least P/R kg/m.(48) 4. As indicated in Fig. LWR/CWR shall not be laid on curves sharper than 440 meters radius both for BG and MG.1 the external equilibrium of a curved elastic beam of radius R subjected to a longitudinal force ‘P’ requires a continuously distributed external force of magnitude ‘f’ where f = P kg / m . 4. This will be derived from the lateral ballast R P is the effective lateral resistance against R resistance τ and τ − buckling danger. . Hence a larger 600mm 100m f COMMON TANGENT POINT l 600mm f 100m REVERSE CURVES NOT SHARPER THAN 20 Note : Shoulder width of 600mm for 100m on either side of common tangent point for reverse curves with radius less than 1500m. Alignment 1.2 : LAYING LWR THROUGH REVERSE CURVES shoulder width on curves and a restriction on the degree of curvature is prescribed. 4.

4. shoulder ballast of 600 mm over a length of 100 m on either side of the common point should be provided.2. 4.4% as laid down in para 419 of IRPWM.(49) 2. These details are shown in Fig. This is because steeper grades imply larger longitudinal forces due to traction and braking which would be detrimental to the health of the LWR causing an increase in the longitudinal stresses in the rail. This vertical curve will serve to smoothen the geometrical transition and introduce a gradual change in the direction of longitudinal force as shown in Fig.3.4. 3. A vertical curve shall be provided at the junction of grades when the algebraic difference between the grades is equal to or more than 4mm per metre or 0. 2. LWR/CWR may be continued through reverse curves not sharper than 875 metres. The steepest grade permitted is 1 in 100 for LWR sections. D & E Minimum Radius 4000 m 3000 m 2500 m Route AII routes Metre Gauge Minimum Radius 2500 m .3. Vertical curves should be of adequate radius as indicated in the table below Broad Gauge Route A B C. Gradients : 1. For reverse curves sharper than 1500 meters radius.

(a). Ballast Cushion and Section : At least 250mm clean ballast stone cushion should be available below the sleeper. The formation width is critical to avoid rolling down of ballast in banks.4. . 4. Yielding formations will result in differential settlement and deformation of track causing misalignment and unevenness.1. a 300mm clean ballast cushion or a 200mm clean ballast cushion with 150 mm of sub-ballast has been prescribed.4.Details of ballast profile in bank and cutting for BG & MG in single and double lines are given in Fig. The ballast section for the LWR is a heaped up section with heaping up of 100mm starting from the edge of the sleeper.α2 (i) Junction of rising or falling grades CHANGE OF GRADE = α1 + α2 (i) Junction of a rising & a falling grade Fig.4. Formation : LWR/CWR should be laid on stable formation.(50) VERTICAL CURVE VERTICAL CURVE α1 α2 α1 l α2 CHANGE OF GRADE = α1 . The shoulder ballast width is increased to 350mm from the standard shoulder width of 300mm. For speeds above 130kmph on BG and 100 kmph on MG.4 TRACK STRUCTURE 4. The stability of the LWR will be affected and buckling of the LWR may result.3 Provision of Vertical curves at grade Intersection Points 4.4. Yielding formations/ troublesome formations should normally be isolated from the LWR section by laying SEJs on either side. In case of curves the shoulder ballast width prescribed is 500 mm. 4.(b).(c) & (d).2.

4. (a) .(51) 30 30 6850 6850 6850 6250 6250 6250 6850 6850 6850 6250 6250 6250 Fig.4.

(52)

30

30

5850 5850 5850 5850 5850 5850

5250 5250 5250 5250 5250 5250

5850 5850 5850

5250 5250 5250

Fig.4.4. (b)

(53)

30

30

12160 12160

11550 11550

12160

11550

12160 12160 12160

11550 11550 11550

12160 12160

11550 11550

12160

11550

Fig.4.4. (c)

(54)

30

30

9810 9810 9810 9810 9810 9810 9810 9810

9210 9210 9210 9210 9210 9210 9210 9210

9810 9810 9810

9210 9210 9210

Fig.4.4. (d)

In double line sections 75% of the keys to be driven in the direction of traffic and 25% in the opposite direction. and any movement which occurs is with rails and sleepers moving together. the breathing lengths shall preferably be provided with elastic fastenings.4.3 Sleeper and fastenings : As explained earlier. In the breathing length. Steel trough sleepers with elastic fastenings for speeds not exceeding 130 kmph (elastic fastenings are used on steel trough sleepers with modified loose jaws or with a steel pad plate welded to the steel sleeper. however.) As an interim measure speeds upto 160 kmph have been permitted with such arrangements. On single line section keys on adjacent sleepers to be driven in opposite directions : 2. Special precautions to be taken for CST-9 track are as under : 1. To achieve this the following types of sleepers and fastenings have been approved for use in LWR/CWR :BG (i) (ii) Concrete sleepers with elastic fastenings. adjacent sleepers will have keys driven in the opposite directions. the entire theory of the LWR is built on the premise that there is no relative slip of the rail with respect to the sleeper. On steel trough sleepers with key fastenings. Steel sleepers with 2 way keys and CST-9 sleepers with 2 way keys are permitted for speeds upto 130 kmph. provided no maintenance problem is faced and performance is satisfactory.(55) 4. Some exceptions to the above recommendations are as below : (i) (ii) (iii) .

5 Direction of key driving with CST-9 sleepers.6 Direction of key driving with steel sleepers. q q .(56) Direction of traffic Breathing length q q Central Portion of double line section Fig 4. Direction of traffic Breathing length Central Portion of double line section Fig 4.

2.6 respectively. MG : The recommended sleepers for speeds above 75kmph but a must for speeds above 100 kmph are : 1. CST-9 sleepers with keys. Sleeper density : The minimum sleeper density (number of sleepers/km) in LWR/ CWR shall be as follows :Type of sleeper i) PRC ii) PRC iii) Other sleepers Sleeper density in BG/MG 1310 in temperature zone I & II 1540 in temperature zone III & IV 1540 in all temperature zones 4. Steel sleepers with two way keys.4 Rails : The following rail sections can be welded to form an LWR : BG 90R 52kg 60kg MG 75R 90R .4. Steel trough sleepers with elastic fastenings. 2. Concrete sleeper with elastic fastenings. Wooden sleepers with anti-creep bearing plates and two way keys or elastic fastening may be permitted. The recommended sleepers for speeds upto 100kmph are: 1.(57) The direction of key driving for CST-9 and steel trough sleepers in LWR territory for single and double line sections is given in Fig 4.5 and 4. if behaving satisfactorily for a maximum speed of 130 kmph in BG and 100 kmph in MG. for continuing LWR.

This makes the behavior of the LWR non uniform. 2. The suggested track structure meets with the requirement of continuing the same track structure for three rail lengths beyond the SEJ.(58) 1.7. In MG. different rail sections are not permitted. eccentricity is induced in the axial forces. 60 R rails converted into LWR may be permitted to continue if showing satisfactory performance. resulting in additional stresses in the rail. (ii) While permitting two different rail sections in an LWR. 3. The destressing temperatures are also different for 52kg and 90R rails.7 . In the same LWR. combination welded joints cannot be avoided.4. As the gauge faces have to be matched. (iii) The ultrasonic flaw detection of combination welds is not completely foolproof.4. following precautions should be taken : 60 KG SEJ 60 kg LWR 60kg 3 Rail panel 52 KG SEJ 52kg 3 rail panel 52kg LWR j combination weld Fig. Before converting an existing fish plated track into LWR/ CWR. This is because of the following reasons: (i) Thermal forces generated in rails of different cross sectional areas are different. The track structure suggested at the junction of a 52kg and 60kg LWRs is shown in Fig.

4. Rails should have a residual life of more than 10 years. 2. iii) 4. Due to this reason. it was considered desirable to isolate the point and crossing assembly from the LWR. Cropping by 300mm – 450mm is generally done. The G3 (L) type glued joint has a pullout capacity of 150 tons for 52 kg rails and 175 tons for 60 kg rails and an insulation .(59) i) ii) Rails to be ultrasonically defective rails replaced. 3. Normally LWRs are not taken through points and crossings. hogged. 4. New rails used in LWR/CWR shall be as far as possible without fish bolt holes. tested and all Rail ends which are bent. Level X-ings should not fall within the breathing lengths of the LWR as level crossings are rigid structures and will not permit thermal movements of the breathing length to take place.5 MISCELLANEOUS 1. However with the introduction of stress frames in the point and crossing area or taking other special measures it is now possible to carry an LWR through station yards including points and crossings. If the rail ends have to be joined during installation then use should be made of 1 m long fishplates/ normal fishplates with screw clamps with SRs of 30 kmph with 1 m long fishplates and 20 kmph with normal fishplates. Insulated Joints for track circuiting in LWR/CWR shall be provided using glued joints of the G3(L) type. battered or having a history of bolt hole cracks should be cropped before welding. For concrete sleeper track a 3-rail panel should be provided between the SEJ and stock rail joint. Similarly a 3-rail panel is required to be provided between the heel of crossing and SEJ. The point and crossing assembly has a distinct track structure with different maintenance schedules and subject to large lateral forces when a train negotiates a turnout.

2. These are :i) The thermal expansion of the deck only in the . SEJ shall also not be located on transition portion of cuves. These interaction effects have been discussed in UIC774-3R (Reference 9) some parts of which have been given below.5. any force or displacement that acts on one of them will induce force in the other. box culverts and arches.5 degree (3500m radius). girder bridges. points and crossings. SEJ with straight tongue and stock rail shall not be located on curves sharper than 0. Glued joints are considered as track components whose procurement has to be done by the engineering department. Bridges with/without ballasted deck with bearings : When the bridge structure and the track exhibit movement relative to each other then there are interaction effects which have to be taken into consideration. LWR on Bridges : (i) Bridges with ballasted decks (without bearing) : LWR/CWR can be continued over bridges without bearings like slabs.1 LWR on bridges: Track Bridge Interaction( UIC 774-3R) 1. gradients and curves. (ii) 4. Given that both track and bridge are connected to one another either directly or through the medium of ballast and are able to move. 4. Location of SEJs : The exact location of the SEJ will be fixed taking into account location of various obligatory points such as level crossings. Introducing a bridge under a CWR track means effectively that the CWR track is resting on a surface subject to deformation and movements hence causing displacement of the track. 4.(60) resistance not less than 25 megaohms in dry condition or 3 kiloohms under saturated conditions.5. All actions which lead to interaction effects are those that cause relative displacement between the track and the deck.

The total support stiffness is composed of the stiffness of each support. ii) iii) iv) v) Horizontal braking and acceleration forces.(61) case of the CWR or the thermal expansion of the deck and of the rail whenever a rail expansion device is present. Effects of temperature gradient. The stiffness K of the support including its foundation to displacement along the longitudinal axis of the bridge is given by . pier. the first 3 are more important. The forces created due to interaction between track and bridge are dependent on a number of parameters of bridge and track both : The bridge parameters affecting the interaction forces are : (1) Expansion length of the bridge(L): For a single span simply supported bridge the expansion length is the span length. base. 3. Rotation of the deck on its supports as a result of the deck bending under vertical traffic loads. Deformation of the concrete structure due to creep and shrinkage. the deck is considered to have two expansion lengths on either side of fixed elastic support. This factor is determined primarily by the total stiffness of the supports. If the fixed elastic support is located at some intermediate point. it is the total length of the deck. For a continuous bridge with a fixed support at the end. foundation and soil. (2) Support stiffness : The resistance of the deck to horizontal displacement is a fundamental parameter as it affects all interaction phenomena. Out of these 5 factors. The stiffness of each support is in turn composed of the stiffness of the bearing.

. ∂p = displacement at the head of the support due to H ∂p H ∂φ ∂h hp φ φ hf Fig.(62) K= H (KN ) ∑ ∂i (cm ) with ∂i = ∂p + ∂φ + ∂h + ∂a where. 4. 3) Bending stiffness of the Deck : As a result of bending of the deck the upper edge of the deck is displaced in the horizontal direction. This deformation also generates interaction forces.8. ∂a = relative displacement between upper and lower parts of the bearing The value of the displacement component is determined at the level of the bearing as shown in Fig 4. displacement due to horizontal movement of the foundation.8 deck’s deformation (this could be calculated assuming the pier to be a cantilever fixed at the base) ∂φ = displacement at the head of the support due to ∂h = foundation rotation.

TRACK PARAMETERS : The resistance ‘k’ of the track per unit length to longitudinal displacement ‘u’ is an important parameter. ballasted or frozen.(63) 4) Height of the Deck : The distance of the upper surface of the deck slab from the neutral axis of the deck and the distance of neutral axis from the centre of rotation of piers affects the interaction phenomena due to bending of the deck.5mm Resistance of rail on sleeper (unloaded condition) k = 40 KN/m u Fig. The value of k has to be established by each railway system as per its track structure. The resistance to longitudinal displacement is higher on loaded track than on unloaded track as can be seen from Fig. the stiffness of the bridge structure and k. 4. standard of maintenance etc.9 TRACK STIFFNESS PARAMETERS (FROZEN BALLAST) Once the values of K. 4. This parameter in turn depends on a large number of factors such as whether the track is loaded or unloaded. use can made of the interaction diagrams given in UIC774-3R for calculation of the additional stresses in the rail and additional forces at the bridge support due to each of the actions causing interaction effects: namely (1)change of temperature (2) acceleration and braking forces (3) deck deformation. the stiffness of the track have been evaluated. .9. Resistance of rail on sleeper (loaded condition) k = 60 KN/m Stiffness ‘k’ u = 0.

3. 4.(64) Resistance of rail on sleeper (loaded condition) k = 40 KN/m Stiffness ‘k’ Resistance of rail on sleeper (unloaded condition) k = 12-20 KN/m u = 2mm u Fig. Due to change of temperature additional stress will develop in the rail and additional force at the support. The design curves for the evaluation of the interaction due to vertical bending of the bridge deck have been evaluated . which in turn causes rotation of the end sections and displacements of the upper edge of the deck. Actions due to braking and accleration The braking and accleration forces applied at the top of the rail are assumed to be distributed over the length under consideration with the following standard values : Accleration = 33 KN/m per track Braking = 20 KN/m per track These values could be modified to take into account the longitudinal loadings given in the Bridge Rules. 2. These are obtained from the interaction charts given in UIC774-3R.9 TRACK STIFFNESS PARAMETERS (NORMAL BALLAST) 1 Changes in temperature : It is assumed that there is change of temperature of ± 35 0 C from the reference temperature for the bridge while for the rail could deviate by ± 50 0 C . Actions due to bending of deck Vertical traffic loads cause the deck to bend.

The maximum permissible additional tensile rail stress is 92 N/ mm2. the global effect ∑ R = αR(∆T ) + βR(braking) + γR(bending) α . γ are the combination factors. The design curves are given for the following two different situations : . the values of the interaction effects can be calculated by using the design graphs in Appendix ‘A’ – page 36 and Appendix ‘B’ page 42 give in UIC report 774 – 3R.(65) with respect to the standard longitudinal plastic shear resistance equal to 20 KN/m and 60 KN/m for unloaded and loaded track respectively. with concrete sleepers and consolidated ballast cushion greater than 30 cm give a total possible value for the increase of rail stresses due to track/bridge interaction as indicated below: The maximum permissible additional compressive rail stress is 72 N/mm2. minimum curve radius 1500 m. For structures consisting of one deck. β . Permissible additional stresses in continuous welded rail on the bridge Theoretical stability calculations on UIC 60kg CWR of a steel grade giving at least 900 N/mm2 strength.through girder bridge – deck neutral axis above track axis. laid on ballasted track. ∑ R is calculated as follows : .deck bridge – the track lies on the top of the bridge deck (deck neutral axis below track axis) . Combining load cases For calculation of the total support reaction and in order to compare the global stress in the rail with the permissible value set by each railway.

This results in an interplay of forces between the girder and the LWR. the magnitude of the force being dependent upon the nature of fastenings being provided between the rail and sleeper. The following cases have been considered: Single span bridge : 1. 2. 4.5. consider the case of a girder bridge provided with fastenings between the rail and sleeper with a creep resistance equal to ‘p’ kg per rail seat. for a simple understanding of the problem let us consider the effect of thermal variation alone as the cause of interaction between the girder and the LWR. other end free.10. To clarify this aspect of interplay of forces between rail and girder. One end fixed. The magnitude of this force developed depends upon the value of ‘p’( the creep resistance) and orientation/nature of the bearings provided in each span of the bridge. Both ends of girder with free bearings. the increment of force in the LWR is np . One end fixed and the other free with similar bearings on a pier 3.(66) 4. As a result of thermal variation the girder has a tendency to expand or contract being provided with bearings. where ‘n’ is the number of sleepers per 4 span with creep resistant fastenings and ‘p’ is the creep resistance . Consequently additional forces are developed both in the girder as well as in the rail. free expansion/contraction of the girder is prevented. One end fixed and the other free with dissimilar bearings on a pier 2.2. The bridge sleepers are rigidly fixed to the top flange of the girder by means of hook bolts. These LWR force diagrams indicate that : i) For sliding bearings at both ends of the girder. On the other hand the central portion of the LWR is fixed in position irrespective of the temperature changes that occur. free bearings at both ends. On variation of temperature due to the creep resistance of the fastenings. The forces developed in the rail and girder for each of the five cases mentioned above are given in Fig. Provisions given in the LWR Manual for carrying LWR over bridges: However. Multiple span bridge: 1.

of sleepers per span p = creep resistance per rail seat Fig.10 (b) . 4.10 (a) →→→ → → → rise n = No.(67) ←←← →→→ rise n = No. of sleepers per span p = creep resistance per rail seat Fig. 4.

of spans n = No. of sleepers per span p = creep resistance per rail seat Fig. 4. 4. of sleepers per span p = creep resistance per rail seat Fig.10 (c) ← ← → →← ←→ → ← ←→ → n = No.(68) → →→ → →→ →→ →→ np p m = No.10 (d) .

the bearings of the two girders are either fixed or free.4. 4. where n = number of sleepers in the span with creep resistance of ‘p’ kg per rail seat (4.10 (e) per rail seat (4. The resultant LWR force 2 diagram is shown in the sketch (4.e. ii) In girders with one end fixed and the other end free the increment of force in the LWR at the roller end is np for a single 2 span bridge.10(c)).10(b)). If it is a multiple span bridge with ‘m’ number of spans. of sleepers per span p = creep resistance per rail seat Fig. the increment of force in the LWR at the roller end will be m×n× p .(69) ←←← ← →→→ ←←← →← np n = No.10(d)). iii) There could be a situation where a pier supports similar nature bearings i. This increment of force will remain the same irrespective of the number of spans of the bridge (4. .10(a)). This is the case when on a pier bearing for one girder is a fixed bearing while the bearing of the other girder is a free bearing. In this case there will be no cumulative build up of force and the resultant LWR force diagram will be as indicated in Fig.10(e).

It is with this assumption that the provisions for laying an LWR over bridges have been framed in the LWR manual. RDSO Report No. which provides an almost zero friction . Fastenings used to connect the rail to the sleeper could be of two types : (1) Creep resistant fastenings and (2) Rail free fastenings which are now termed as zero longitudinal restraint fastenings. In case of large lateral forces. The pad under the rail is made up of low friction material like teflon. 4.(70) Fig. C-169 investigates the creep resistance offered by different types of rail –sleeper fastenings. the baseplate prevents the overturning of the rail. On the Indian Railways we have been traditionally using dog spikes and rail screws as rail free fastenings although now Pandrol has come up with a zero longitudinal restraint design(Fig 4. Under normal circumstances there is a small gap between the base plate (steel) and the top side of the rail foot.11 In order to avoid interplay of forces between the LWR and girder a possible solution would be to provide rail free fastenings between rail and sleeper on the girder bridge.11).

(71)

F B B’ C’ C

A

F

D

Fig. 4.12

F

L0 B C F G

A

D

E

H

Fig. 4.13

movement between the rail and sleeper. Use of rail free fastenings on bridges where LWR is proposed to be used is now mandatory due to requirement of minimizing the interaction of forces between the LWR and the girder. However, this results in another problem : enhanced gap at fracture, when the fracture occurs on the approach of bridge laid with LWR. Consider an LWR laid on normal formation with the usual force diagram A B C D. in the event of fracture at location ‘F’ the stress in the LWR is released at that location and two new breathing lengths B1F and C1F are formed on either side of the fracture location. (Fig 4.12)

(72) The gap g1 at the fracture location will be given by

AE (αt ) × 2 ———— (1) g1 = 2R|
2

[Assuming equal movement on either side of F] R| represents the longitudinal ballast resistance mobilised at the time of the fracture, which is generally about 50% to 60% of the normal R value, due to the sudden nature of occurrence of a fracture. However, if the same fracture had occurred in the approach of a bridge provided with LWR and rail free fastenings the modification of the force diagram will be as given in the figure 4.13. In this figure, ABCDEFGH represents the altered force diagram. Gap at fracture in this case will be

2xAE(αt ) + L0 .α.t ————— (2) g2 = 2R|
2

Where L0 is the span length of the bridge provided with rail free fastenings. Expressions (1) and (2) indicate that the gap at fracture is enhanced by an amount equal to

L0αt , when a girder bridge with

rail free fastenings is located in the central portion of the LWR. Indian Railways have fixed the permissible gap at fracture as 50mm where by expression (2) becomes

2x

AE(αt ) + L0αt < 50mm 2R|
2

This expression is applicable for both BG and MG tracks. However, as the wheel diameter of MG stock is smaller than

(73) BG, the fracture gap of 50 mm is more critical for MG. Over the years attempts have been made to increase the value of L0 by adopting various techniques :(1) One way could be to increase the value of R, the longitudinal ballast resistance mobilized at the fracture. This could be done by :·l Compacting the ballast in shoulders and cribs of the bridge approach sleepers. ·l Enhancing the sleeper density to 1660 Nos./km in the bridge approach. ·l Heaping up of ballast in the bridge approach starting from the foot of the rail. ·l Box anchoring sleepers wherever required. These measures have to be taken in the bridge approaches 50m on either side. Table 1 of the LWR Manual 1996 gives the maximum overall length of girder permitted on LWR/CWR in with the following stipulations : 1. Girder bridge should have sliding bearings on each end with single span limited to 30.5m. 2. Rail should be provided with rail free fastenings throughout the length of the bridge from abutment to abutment. 3. The approach track should be suitably upgraded as mentioned above. 2) Another way of increasing the value of Lo would be to improve the approaches as mentioned above in addition to providing a few sleepers on each span with creep resistant fastenings. The creep resistant fastening will hold the rail and prevent the gap at fracture from becoming excessive. However, provision of creep resistant anchors implies an interplay of forces between the rail and grider. Hence the following stipulations are made for bridge provided with rail free fastenings and partly boxanchored (with single span not exceeding 30.5m and having sliding bearings at both ends).

These are discussed below : (1) Providing an SEJ on each pier with rail free fastenings on the bridge. if the girder is having rollers at one and rockers on the other side.4. Table 1 of LWR Manual Temperature Zone Rail Section Rail free fastening Rail free fastening on bridge on bridge and partly box-anchored/ creep resistant fastenings Type of sleeper Type of sleeper used in approaches used in approaches I II III IV 60kg 52kg/90R 60kg 52kg/90R 60kg 52kg/90R 60kg 52kg/90R PRC/ST 30m 45m 11m 27m 11m 27m 11m 27m PRC/ST 77m 90m 42m 58m 28m 43m 23m 43m The LWR Manual has also suggested some additional methods of carrying an LWR over bridges. (2) Bridge timbers laid on girders shall not be provided with through notch but shall be notched to accommodate the individual rivet heads.14. (4) The sliding bearings shall be inspected twice a year and oiling and greasing of the bearing carried out once in two years.(74) (1) On each span 4 central sleepers will be provided with creep resistant fastenings and remaining sleepers with rail free fastenings. This arrangement is shown in Fig. (3) The girders shall be centralized with reference to the location strips on the bearing before laying LWR/CWR. These sleepers will be at the centre of the span if the girders are having sliding bearings on both sides. These sleepers will be at the fixed end of the girder. . In order to avoid creep four sleepers on each span will be box-anchored.

The SEJ will have to cater to the free expansion or contraction of the rail on the bridge as well as movement of the breathing length. .Hence the SEJ will have to be a wide gap SEJ capable of accomodating larger movements.The permissible span lengths with normal SEJs and 190mm maximum gap SEJs are given on the adjoining page.(75) (2) Providing an SEJ at the far end approach of the bridge using rail free fastenings over the girder bridge(Fig 4.15): In this arrangement an SEJ is provided at the far end approach of the bridge(abutment away from the LWR) at a distance of 10m away from the abutment with rail free fastenings on the bridge proper.

move ment of SEJ (mm) 190 190 190 190 120 120 20m 50m 15m 50m 4.0 cm 4.0 cm 55m 70m 110m 160m 45m 70m 100m 150m 7.5 cm 6.0 cm 7. .0 cm II I Note : SEJ is to be installed 10 metre away from the abutments.0 cm 6.0 cm 4.0 cm 4.5 cm 6.Max. Zone (76) IV III II I 6.5 cm 6. length of bridge with SEJ Max.5 cm with ST/ PRC approach sleepers with CST-9 approach sleepers with ST/ PRC approach sleepers Initial gap to be provided at td with CST-9 approach sleepers Rail temp.5 cm 6.

14 Fig. 4.(77) Fig. 4.15 .

A foot by foot survey is recommended. need to be identified. 4. 3. points and crossings. yards.H.(78) CHAPTER V LAYING AND MAINTENANCE 5. The maximum and minimum rail temperature for a continuous period of at least 5 years shall be ascertained and the mean rail temperature for the region arrived at. An initial survey of the section where the LWR/CWR is proposed to be laid should be carried out. This plan should be got approved by the T.1 Laying of LWR 1.O. This could provide the basis for fixing the rail neutral temperature. ascertaining the periods during the year when . Locations where LWR/CWR cannot be laid. A detailed plan shall be made showing the exact location of the SEJs and various other features such as level crossings. These locations could be: i) Sharp curves ii) Bridge locations iii) Steep Gradients iv) Points and crossings v) Troublesome formations vi) Distressed bridges These locations will be isolated from the LWR by providing SEJs on either side. if there are no deviations from the provisions of the LWR manual or by CE/CTE if any deviations are proposed. 2. gradients and bridges. Temperature Records : The LWR Manual (1996) prescribes that each railway should nominate 8 to 10 stations on its jurisdiction where temperature records over a period of 5 years should be built up by installing suitable continuous recording thermometers.D. curves. This is called the LWR plan.

5 metres or longer rail pieces of the same rail section as the LWR. 10-rail or 20-rail panels. Destressing equipment wooden mallets. Rail cutting equipment. 5. Care should be taken while handling 90 UTS rails at the time of unloading. Stabilization of troublesome formation. use may be made of the rail temperature map given in the LWR Manual. Realignment of curves. roller. The 10-rail/15-rail/20-rail panels should be placed in the track and subsequently welded to form the LWR of required length. Welding of rails to form LWR : ii) . Introduction and improvement of vertical curves. Preliminary works to be carried out prior to laying the LWR : i) ii) iii) iv) v) vi) i) Replacement of insulated joints by glued joints if LWRs are to be laid in station yards. 7. Rehabilitation of weak bridges. If these records are not available. 10-rail. ii) iii) 2 sets of SEJs with sleepers and fastenings. Rails are supplied from workshop flash butt welding plants in 3-rail. iv) v) vi) vii) 6. i.e.(79) maintenance operations could be carried out or hot weather and cold weather patrolling need to be introduced. tensor. Adequate number of 1 meter long fishplates with screw clamps/joggled fish plates with slotted grooves and bolted clamps. Lifting or lowering of track to eliminate sags and humps. Materials required for laying an LWR : i) Four numbers of 6. Rail closures. Welding equipment.

long fishplate with screw clamps or a joggled fishplate/normal fishplate with screw clamps exists. Closure rails of 6.(80) iii) Two complete sets of SEJs shall be inserted at predetermined locations with gaps in the mean position. Speed restrictions (30 kmph) should be imposed if fishplated joints are existing in an LWR. Destressing is the operation of removing the locked up stresses in the LWR and bringing the LWR to a stress-free state at a predecided temperature called the stressfree temperature. A watchman should also be posted at a clamped joint. The LWR Manual lays down that destressing should be done : . If a temporary joint in the form of a 1m. Subsequently these instructions were withdrawn and current instructions do not lay down any periodicity for destressing. destressing of the LWR should be undertaken as soon as possible. Laying of welded panels or welding of joints at site can be done at any time of the year.1 One of the most important maintenance operations in the LWR is destressing the LWR. 5.2 Periodicity of destressing and conditions which would warrant destressing to be done : Following the Khanna accident. a periodicity of once in 3 years for destressing every LWR had been prescribed.5 metres or longer length shall be provided on either side of SEJs to facilitate adjustment of gaps during destressing operation. iv) v) 5. then speed restrictions of 30 kmph and 20 kmph respectively should be imposed. As per the LWR Manual.2 Destressing Operations 5.2. the stress-free temperature should lie between tm + 50C and tm + 100C for 60kg and 52kg rail sections and between tm and tm + 50C for 90 R rail sections where tm denotes the mean rail temperature. But after welding sufficiently long panels of about 1 km length or longer.2. It is also called the neutral temperature.

2.Block Activities: 1) 2) 3) 4) All impediments to free rail movement such as check rails. 4 Nos. 4. After a special maintenance operation like deep screening. next to the SEJs.3 Destressing operations could be done in two ways : (1) Manually (2) Using a tensor 1. should be removed. 5. After restoration of track following an unusual occurence.(81) 1. B) Block Activities: 1) . ERC clips should be greased so that their removal during the block will be easy.2. Prior to the block. rail anchors etc. If the number of locations where temporary repairs have been done exceed 3 per kilometer. 50% of the ERCs are loosened and a speed restriction of 30 kmph imposed. Arrange for block of adequate duration taking into account the length of the LWR to be destressed and the labour available. Work should be done in the presence of a PWI. (i) when gap at SEJ goes beyond the prescribed limits. It is suggested that the condition of the entire LWR should be considered before taking a decision of destressing. 3. The steps involved are given below : A) Pre. (ii) when tongue rail/ stock rail cross the mean position. Manual Destressing of the LWR : When the prevailing temperature tp falls in the range prescribed for the stress-free temperature then manual destressing of the LWR can be resorted to. of closure rails to be created at either end of the LWR.

the rail tensor can be used for destressing operations.(82) 2) 3) The closure rails are removed and placed on the cess at either end of the LWR. Welding of the closure rails can be performed if required in a separate block. Principle : The rail tensor creates tensile stress in the rail of such magnitude that these stresses completely balance the compressive stress created in the LWR when the temperature rises from tP (the prevailing rail temperature) to tn. The rails are tapped with a wooden mallet to remove any builtup stress and subsequently lowered on the rail seat after removing the rollers. This is a good opportunity to change the rubber pads. The SEJs at either end are adjusted for a standard 40 mm gap at the destressing temperature. Side rollers should be used for curves. ERCs are put back starting from the middle towards the ends.1 shows schematically the operations involved in destressing manually with the options available. The rail tensor is a hydraulic or mechanical device which can create tensile stress in the rail by pulling the same. Fig 5. Suitable cuts in the closure rails can be made and inserted back in the track after which the block can be removed. 4) 5) 6) 7) 2. The rails are subsequently lifted off the rail seat and supported on rollers at every fifteenth sleeper. During this operation the temperature of the rail should lie within the prescribed range for the destressing or stressfree temperature. ERCs are removed starting from the ends towards the centre. Destressing using tensor : When the prescribed temperature range for the neutral temperature is not available in the prevailing conditions. the defined rail .

(83) neutral temperature. If the length of the section is L. The LWR is made stress-free at tp by removing the fastenings and subsequently tensile stresses are created in it by pulling the LWR by a calculated amount : Compressive stress created in the LWR due to rise of temp from tp to tn with zero stress at tP = E α (tn – tp) The tensor creates the same amount of tensile stress by pulling the LWR section by an amount equal to ∆ . enabling trains to pass when in position. It can be dismantled into different parts for being transported. then strain = E ∆ L ∆ L The tensile stress due to this strain will be ∴E ∆ = E α (t n − t p ) L or ∆ = Lα (t n − t p ) This expression gives the extension to be imparted to a segment of length ‘L” of the LWR to get the stress-free temperature at tn. dipped joints and rail-head marks. Pune are given below: Total weight = 325 kg (without pump) Pulling force = 70 tonnes Pushing force = 30 tonnes Stroke = 380 mm Overall dimensions : 4500 mm length 1200 mm width . (Details of a tensor are given in Fig 5. the weight of the heaviest component being 54 kg. It grips the web of the rail using a special cam arrangement eliminating bending. It is non-infringing.2) Some of the details of a tensor manufactured by CTR Industries. Equipment : The tensor is basically a pulling device which could be hydraulic or mechanical.

3. 2. 5.1(b) Destressing Manually By Splitting Long LWRs. Level crossing Note : 1.7. 3. Fastenings removed from ends towards centre.1(a) Destressing Manually 3. Cuts made as indicated prior to block. .6.5. LWR has been split into three convenient sections. Level crossing used as a convenient location to split LWR.8 to create two pairs of closure rails. 2. 2. Rail cuts at locations 1.(84) Note : 1.4. 5. Fastenings put-back starting from centre towards ends. Fig. Fig.

4) Unfasten the fastenings starting from the end or from the centre where the rail is cut..(85) Steps in destressing by the rail tensor (Fig 5.. W1. 3) Erect marker pillars W0.2. An estimate of the anchor length is as under: For BG For MG .5 m per C of t -t n p This anchor length is excluded from the destressing operations and will have to be done manually. The first marker pillar W0 will be erected at a distance equal to the anchor length from the SEJ.tp) . The anchor length is the length of track where the fastenings will not be removed during the destressing operations. Note movement at W0. Place the rail on rollers. = Y0 + (W0 W1)x α x(tn . Transfer W0 on to the foot of the rail. The last marker pillar will be erected at the point where the LWR is being cut to make a gap of 1m. Wa on to foot of rail and note down the prevailing rail temperature tp. W2 at convenient distances of say 100m. Calculate the extentions to be given in each segment. W2. . This length is required to resist the pull applied by the tensor. If the movement is away from tensor.... the zero correction is positive..4. 2) Make cuts 1m apart at the centre of the LWR. Extention to be given to first segment WoW1 will be. Let it be Y0 5) Transfer the remaining points W1.5 m per C of t -t n o o p . Apply pull by tensor to get some movement at W0 and release. These marker pillars may be made on both the sides of the track..3) : 1) The destressing operations have to be carried out when the prevailing rail temperature tp is less than the the designed neutral temperature..

9. FORK END 4 12. ITEM No. 1. 2. 5. CLEVIS PIN WITH CHAIN HOSE PIPE-2m 10. JAW TIE BAR 4 2 QTY. 3. HYDRAULIC PUMP UNIT 6. 7. ITEM No.(86) 10 11 12 9 3 1 2 4 6 8 5 PART LIST Sr. T CONNECTOR 2 2 1 QTY. 2 4 2 Sr. 2 6 4 7 Fig. Nos.2 HYDRAULIC RAIL TENSOR . HOSE PIPE-4m 11. YOKE LEVER ARM HYDRAULIC CYLINDER 4. 5. 8. No.

(87) tn Fig.3 . 5.

. 5) Renewal of fastenings requiring lifting of rail. otherwise use the tensor to make the gap 25mm and do the welding. extention to be given to segment W1W2 = Extention to segment W0W1 + W1W2 x α x (tn-tp) This distance is marked at W2 away from the tensor. 6) 7) 8) 9) 5. The other gap could be welded also. if 25mm. In another block make a cut at the earlier made paint mark and put in a closure rail piece of 6. At the end in order to replace the 1 metre long closure rail at the centre make a paint mark at a distance of (6.3 5.This can ensured by bringing the mark of required extension opposite to the mark on the marker pillar W1. Remove the tensor and normalize the block. One gap can be welded immediately.(88) This distance is marked on the foot of the rail at W1 away from the tensor.3.1. Similarly.5m + 2 welds) measured from one rail end across the tensor. 6) Maintenance of SEJ/buffer rails.5m. Regular maintenance operations Regular track maintenance operations as defined by the LWR Manual are : 1) Tamping/Packing 2) Lifting 3) Aligning including minor curve realignment 4) Shallow screening/ shoulder screening. This process could be continued on to the other side of the tensor as well either simultaneously or after tackling one side of the LWR. Once the required extension has been given to a segment the segment can be lowered on to the rail seat and fastenings put back. Apply pull by tensor to get the required extention of segment WoW1.

For track structure consisting of concrete sleepers : Passage of 50. 5. last two of which should be with on track tamping machines.4 Temperature Restrictions while Maintenance Operations in LWR territory : (i) carrying out Normal maintenance operations should be carried out well before on set of summer.3. Most of the operations involve disturbance of the ballast bed leading to loss of the sleeper to ballast resistance. (ii) If the rail temperature rises above td + 200C during the period of consolidations which has been defined earlier (para 1.18 of LWR Manual) then the following steps have to be taken : .00.000 gross tons of traffic on BG or at least 1. III For newly laid LWR/CWR at least three rounds of packing.000 gross tons of traffic on MG or a period of 2 days whichever is later. I (i) (a) (b) (ii) II One round of stabilization by Dynamic Track Stabiliser (DTS).000 gross tons of traffic on MG when compaction of ballast is done using handoperated compactors/ consolidators or rammers. Consolidation of track is the process of building up of sleeper to ballast resistance either initially before laying of LWR or making up subsequent loss of resistance by any one of the following (Para 1.2.(89) 5. Passage of at least 50.000 gross tons of traffic on BG or at least 20. when compaction is done by mechanized shoulder and crib compactor. The temperature range within which such operations should be performed should be restricted to td+100C to td–300C where td is destressing temperature.3.00.18 of LWR Manual) For track structures consisting of sleepers other than concrete sleeper : Passage of al least 3.000 gross tons of traffic on MG or a period of 2 days whichever is later.000 gross tons of traffic on BG or 20.

A speed restriction of 30 kmph in BG and 20 kmph in MG will have to be imposed if mechanical compaction of ballast has not been done during the maintenance operations in addition to posting of a mobile watchman. Manual Through-packing of sleepers in LWR territory : It must be remembered that till off track tampers are made available manual packing of concrete sleeper track with crowbars has been permitted (para 1408 of IRPWM). Other instructions while carrying out track maintenance in LWR territory: (i) Special attention shall be paid to . A speed restriction of 50 kmph in BG and 40 kmph in MG will have to be imposed. (B) 5.5. for undertaking manual packing.3.6. A speed restriction of 50 kmph in BG and 40 kmph in MG will have to be imposed if shoulder and crib compaction has been done during the maintenance operations. ii) 5. 2. leaving the next 30 sleepers fully boxed and packed. However. The intervening sleepers can be opened out for packing after 24 hours if the GMT on the BG section is more than 10 and after 48 hours for BG sections with GMT less than 10 or in MG sections. Only in case of emergencies and in the presence of PWI can 100 sleeper spaces be simultaneously opened out observing the usual temperature restrictions. the following precautions have to be observed: i) Only 30 sleepers should be opened out at a time.(90) (A) For other than concrete sleeper track : 1. For concrete sleeper track : 1. due to the necessity of opening out of ballast.3.

points & crossings and unballasted deck bridges. Horizontal and vertical curves. Maintenance tamping using track machines can be done in continuation from one end of the section to the other. (ii) Special attention shall be paid to maintenance of fastenings in LWR/CWR especially on concrete sleepers . In order to prevent rolling of ballast down the slope. Ballast procurement being a long lead item. pedestrian and cattle crossings.3. bridge approaches. adequate planning should be made for timely procurement of this vital item. Ballast section shall be properly maintained paying special attention to locations where the ballast profile could get disturbed such as at level crossing approaches. Ballast should be recouped well before the onset of summer.7 Mechanised Track Maintenance : (i) . Dwarf walls should be provided wherever track trespass is anticipated to prevent loss of ballast. General lift should not exceed 50 mm for concrete sleepers and 25 mm for other than concrete sleepers. the cess should be properly maintained with regular cess repairs. (iii) (iv) (v) 5.(91) maintenance of track at the following locations : SEJs/breathing lengths. care should be taken to avoid simultaneous lifting of track. Approaches of level crossings. While slewing of track using crowbars. curves. Adequate ballast should be arranged before going in for any maintenance operation such as lifting.

rails and fastenings : The following precautions should be observed : (i) Casual renewal of sleepers : Not more than one sleeper in 30 sleepers shall be replaced at a time.3. in excess of 50 mm for concrete sleepers and 25 mm for other sleepers shall be carried out in stages with adequate time gap in between the successive stages. such that full consolidation of the previous stage is achieved prior to taking up the subsequent lift. Casual renewal of fastenings : a. Fastening renewal requiring lifting of the rail such as replacement of the grooved rubber pads shall be done in the presence of the gang mate with at least 30 sleepers in between to be kept intact. b. packed and aligned once in a fortnight. Should it be necessary to renew two or more consecutive sleepers in the same length. Work shall be done under the supervision of a keyman. Maintenance of SEJs and buffer rails : i) SEJs should be checked. fastenings of not more than one sleeper at a time shall be renewed at a time. Lifting of track where needed. while at least 15 sleepers in between shall be kept intact.3. Oiling and greasing of the (ii) 5.(92) Rail temperature restrictions are the same as laid down for manual track maintenance. 5.9. When fastening renewal does not require the rail to be lifted. they may be renewed one at a time after packing the sleeper renewed earlier duly observing the temperature limits.8 (ii) . Casual renewal of sleepers.

Movement of rail over sleeper in breathing length. (c) In zone III and IV. 4 buffer rail pairs shall be provided. In zone I & II the lower temperature limit for gap fully opening out and the upper temperature limit for gap fully closiing shall o be tacken as td0 25 Cand td+25 C respectively. the LWR should be destressed and the gap at the buffer rail joints restored to 7. if the gap closes at a temperature lower than td + 300C or opens out to 15 mm at a temperature higher than td – 300 C it would indicate : i) ii) iii) iv) Defective initial gaps.0 meters. Creep of LWR. (a) In rail temperature Zone I and II. 3 Buffer rail pairs. (b) Buffer rail joints are lubricated twice a year when the rail temperature is between td + 150C and td – 150C. the buffer rail shall be 6.5 mm is provided at a buffer rail joint at temperature td. A standard gap of 7. while in zone III and IV.(93) tongue rail and stock rail should be done simultaneously. ii) Buffer rails are free rails placed in lieu of Switch Expansion Joints. This would include the following items : . 5. In such cases. On BG. Inadequate packing in breathing length.5 m long while for MG it will be 6.5 mm at td.4 Special Track Maintenance : These works are generally carried out with a speed restriction in force.

Major lowering/ lifting of track Major realignment of curves. 2. Work will be done with an SR of 20 kmph in the presence of a PWI. there should be at least 4 intermediate sleepers fully packed and boxed. 2. Sleeper renewal renewal. other than casual Formation rehabilitation. This implies that 1. If there is a possibility of rail temperature rising above td + 100C during the execution of work then temporary destressing at a temperature 100C below the anticipated maximum temperature should be carried out. cold weather patrolling should be introduced. Deep screening in LWR territory : Provisions laid down in Para 238 of IRPWM will also apply mutatis mutandis to LWR/CWR.(94) i) ii) iii) iv) v) vi) Through fitting renewal Deep screening/ Mechanised cleaning of ballast. While tackling two sleepers simultaneously. 3. Temporary destressing days if there is wide there is a possibility beyond the anticipated should be done again after 15 fluctuation of temperature and of temperature rising further maximum temperature. The temperature restrictions are as under : 1. 5. This will keep the track in the safe zone as far as development of high compressive forces is concerned. . 3. If the temperature were to fall by more than 300C below the temporary destressing temperature. Deep screening could be carried out in continuation from one end of the section to the other in the above manner. Work should normally be done in the rail temperature range of td– 200C to td + 100C.5.

.(95) 4. Once the deep screening is completed. then the entire LWR should be destressed to bring the destressing temperature to the normal range.

(96) CHAPTER VI UNUSUAL OCCURENCES IN LWR. fractures could occur in a rail due to variety of causes : 1.2. scoring. plastic deformation. 3. rolling.2 Fractures 6. . Dynamic stresses caused by vertical and lateral loads particularly by vehicles with wheel flats or when the vehicle runs over poorly maintained rail joints etc.1 Rail and weld fractures occur with increasing frequency on LWR sections in winter due to development of longitudinal tensile forces as rail temperatures fall below the destressing temperature. guide marks etc. hydrogen flakes. Defects occuring due to incorrect handling of rails e. INSPECTION & RECORD KEEPING 6. min Causes of fracture : Apart from excessive tensile thermal forces which could arise in a rail in LWR sections. 6. Residual stresses induced during manufacture : cooling. Material defects originating during the manufacturing process such as clusters of non–metallic inclusions. straightening etc. 4.1 Introduction: This chapter is devoted to a discussion on various unusual occurrences which may occur in an LWR such as fractures and buckling. which may be present in spite of the non–destructive tests carried out on the rails during their quality assurance examination. rolling marks.g. The increasing incidence of rail fractures in LWR sections could also be attributed to the fact that the rail distressing temperature has been fixed between tm + 50C to tm + 100C for 52 kg and 60 kg rails increasing the tensile force created in the rail as temperature drops towards t . and remedial measures to be taken in the event of the same. 5. denting etc. 2. Defects associated with faulty welding.

the train is allowed to move over the joint with a SR of stop dead and 10 kmph for the first train and 20 kmph for the subsequent trains. 2. Fractures due to corrosion at rail seat and liner location etc. a fishplate for a rail fracture or a joggled fishplate for a weld fracture is fixed.2.5 m. If one meter long fish plates are used during the repair the SR will be 30 kmph.2 Repairs to Fractures 1. fracture repairs are done in the following stages : A. If the gap at fracture is more than 30 mm. The LWR Manual authorises a keyman / gangman to pass the train in such an emergency. B. using 4 tight screw clamps without a rail closure piece. Two paint marks are made on either side of the fractured . Emergency Repairs These repairs are carried out to pass the train over the fractured rail. 6. Temporary Repairs This essentially involves the removal of the fractured rail from the track and replacing it with a sound rail–closure piece of length generally more than 6. a rail–closure piece will be inserted into the gap after which fishplates or joggled fishplates will be fixed. 2. If the gap at the fracture is less than 30 mm. 3. Once a fracture has occurred. The following steps are involved : 1. 4. This done. 1. As discussed earlier excessive tensile forces in the rail generated due to temperature changes of the LWR. If necessary. The following steps are involved in temporary repairs. a wooden block may be inserted below the rail to support the fractured joint. 7. This done. the railway official detecting the fracture should take immediate steps to block the section and prevent any train movement over the fractured portion.(97) 6.

rail cuts are made on either side of the fractured joint at the paint marks made earlier. Let the length of this closure rail piece be L. 3. These distances X and Y are correlated to the length of the closure rail piece which is to be inserted into the track. Fig 6. say X and Y as shown in the Fig 6.1 2. Here length of rail inserted = L + 2 welds (50mm) Length of rail removed = X + Y + 2 saw cuts (say 1mm ) Hence L + 2 welds (50 mm) = X + Y + 2 saw cuts (say 1mm) This relationship enables fixing up of paint marks on either side of the fractured rail joint at distances of X and Y from the fracture location.(98) joint at a distance. the principle to be observed is that the rail inserted into the track should be equal to the length of the rail removed. The closure rail piece of length ‘L’ is inserted into the gap created and fishplates with screw clamps fixed at the two joints. During a block of adequate duration. While replacing the fractured rail. .1.

(wide gap weld) Rail removed = X+Y+2 saw cuts (1mm) 75 = X+Y+2 saw cuts . To summarise. 5. It will be used for reducing the gap to 25 mm. and fastenings removed over a length of 125 m on either side of the weld. as will generally be the case. A. 6. if the other gap is not 25 mm. If the gap at fracture is bridged by providing a longer length closure rail. The rail is tapped to equalise the stresses and fastenings put back in position. there will be a drop in the neutral temperature. For welding the other gap. However. In the same block if time permits. or in another block one of the gaps is adjusted to 25 mm (for SKV welding) and welding performed. This is done by removing sleeper fastenings over a given length and applying the requisite pull by tensor. 7. With the tensor in position and gripping the rail web.2. the rail joint is welded. a tensor is used. If the other gap is also 25 mm then the other joint could also be welded. The method of repairs has been indicated in Fig.B are paint marks on either side of fracture. A new development in welding technology is wide gap welding. This enables fracture repairs to be done by providing a single weld instead of two welds as was done earlier. The tensor is kept in position till weld metal cools down. 6. The temporary repairs should be carried out in the supervision of a PW Mistry/PWI.(99) 4. the process of temporary repairs is carried out without adding any additional rail metal to the LWR. creating high compressive stress in the rail during the summer season. Rail inserted = 75mm. After the weld metal has cooled down. then the other gap is fishplated with screw clamps and opened to traffic. the tensor is removed.

6. Initially it was thought that the slender rail section would not be able to take the high compressive forces generated during the summer season. 3. Buckling results in complete distortion of the track geometry affecting safety and it is not possible to pass a train over the buckled track. Tensor for obtaining the standard gap. Rail Closure Pieces with different lengths. 2.3. 4. 5. Subsequent investigations by . Fishplates/ Joggled fishplates with bolted clamps.2 C. The Manual prescribes destressing to be done when the number of fractures exceeds 3 per km.(100) LOCATION OF FRACTURE Fig.1 Buckling is the phenomena describing the sudden lateral shift in the alignment of an LWR to relieve the built–up compressive forces during the summer months as the temperature rises above the destressing temperature.3 BUCKLING : 6. lateral stability of LWR track in hot weather conditions has been a source of great concern to track engineers. Permanent Repairs : This will involve destressing the entire LWR after a number of fractures have occurred in the LWR. One metre long fishplates with bolted clamps. 6. D. Equipments required for fracture repairs : 1. Over the years. Welding equipment with all accessories.

5. Improper functioning of the SEJ. Missing fittings. 4. the following steps are suggested : 1. however. avoiding addition of metal at the fracture location during repairs.e.3 Steps to be taken to avoid buckling : 1. 7. Non–observance of the specified temperature restrictions while performing maintenance operations in an LWR. Full complement of fastenings and anchors. 5. 4.3. 6. Ensuring proper ballast profile. Introduction of hot weather patrolling when the prevailing 0 rail temperature goes beyond td + 200C. 2. indicate that the track strength against buckling was contributed by not only the rails. depending upon the severity of the situation. 3.2 Some of the factors which could lead to buckling are : 1. The section to be blocked or speed restrictions to be imposed. . 6. (tp>td+20 C). by taking out. Keeping a close watch on SEJ gaps specially during extreme temperatures. 3. Controlling misalignments in track. if required. and the ballast contributed in a substantial measure to the strength against buckling. 2.3.(101) various railways. but the rail sleeper assembly with fastenings. Proper repairs of fractures i. 6. Additional ballast to be dumped on the shoulder.4 Steps to be taken in face of impending buckling : On detecting severe sunkinks or noticing hollowness of sleepers as detected by a canne–boule. 6.3. Observance of specified temperature restrictions (tp < td + 100C) during maintenance operations. ballast from the centre of the track. Settling formation resulting in poor alignment of track. Lack of ballast affecting the lateral and longitudinal ballast resistances. 2.

In the morning times.1 Inspection: While an LWR section. The golden principle to be followed is : “When in doubt.(102) 3.” 6. In order to complete the repairs. However. A closure rail piece of suitable length could now be inserted into the track and section restored after fixing fishplates and screw–clamps.4. After removal of the rail piece. AEN & PWI and other inspecting officials should pay special attention to the aspects given below while inspecting LWR sections. 6. cut rail out. rail should be cut out from the track. . DEN / DEN.5 Repairs to be undertaken in the event of buckling : 1. destressing of the entire LWR will be carried out as early as possible. this could be achieved using a hacksaw blade. Subsequently the heat affected martensite zones could be removed by cutting three inches on either side of the gas cut by a hacksaw blade. At tight rail locations.4 Inspection of the LWR and Record keeping : 6. (B) Permanent Repairs The clamped closure rail piece will be welded at either end. The rectification shall normally be carried out in the following stages under supervision of the PWI. (A) Emergency Repairs 1. A 6.3. The Sr.5 m long rail piece will be cut out from the buckled track at the location of buckling. rail cutting equipment will be required. This repair is carried out to restore traffic on the section. Each case of buckling shall be investigated by the AEN soon after its occurrence and a detailed report submitted to the DEN/Sr DEN. 2. 3. Gas cutting to cut rail out will have to be resorted to. To get the required gaps for welding. reduces the maintenance requirements it necessitates intensive inspections at supervisory and officers' level. resorting to gas–cutting if required. 2. this will not be possible when the rail is under compression as it will tend to pinch the blade. it will be possible to slew the track back into proper alignment.

girder bridge approaches. destressing . Keyman and Gangman in the event of a fracture or buckle. (i) (ii) 6. Gangmen and P Way Mistries for track maintenance in LWR territory. 2. Keymen. Mate. Special attention at vulnerable locations such as curves. (2) An indication plate should be fixed on the cess at each SEJ. The sectional AEN will measure the SEJ gaps once in 6 months preferably during the coldest and hottest months. Inspection of SEJ gaps and creep movement in central portion of LWR / CWR should be as per schedule given below: PWI / APWI To measure the SEJ gaps alternately once in 15 days during the two hottest and two coldest months of the year. 4. 6.4. This register should record various details of the LWR as laid down in annexure XI and XII of the LWR Manual. showing the date of destressing. level crossings.. Ultrasonic Flaw Detection of rails and welds should not be in arrears.(103) 1. 3.2 Records: (1) The PWI should maintain a permanent register called the LWR section register. Ballast adequacy and maintenance of stipulated ballast profile specially at locations where the profile is likely to be disturbed due to trespass. Action to be taken by P. etc. they will measure at 2 monthly intervals again alternately. Way Mistry. In the remaining 8 months. 5. Knowledge of rules and regulations (specially temperature restrictions) of Mates.

responsibilites and staff training Readers may refer to para 9 of LWR Manual 1996 to know more about the duties. Inspection of SEJ gaps and creep movement in the central portion will be recorded as per laid down frequency in the proforma prescribed in AnnexureXIII(A) and XIII(B). initial each page and send exception report to Territorial Chief Engineer for his decision / orders.(104) temperature and length of LWR / CWR. . responsibilites and training of staff for working in LWR territory. DEN / Sr. (3) (4) (5) 6. PWI / AEN / DEN will carefully study the SEJ gaps and creep in the central portion before deciding the remedial measures required to be taken.DEN will scrutinize observations of each LWR /CWR.5 Duties. AEN will analyse the observation of each LWR / CWR in his jurisdiction and give a certificate at the end of the LWR / CWR section register before onset of summer regarding satisfactory behaviour of all LWRs / CWRs on his section.

The following results were obtained from the tests : 1) In all the tests the track buckled laterally. Buckling of straight tracks occurred suddenly with a loud bang (explosive buckling) while the imperfect track buckled gradually and quietly(passive buckling). Straight tracks with smaller lateral imperfections buckled at much higher temperature increases than those tracks with noticeable lateral imperfections. The track section was 46.(Fig 7.3 or 4 noticeable half waves each of length 5 to 6 metres. This implied that a buckled track could have several shapes with buckling taking place in several wave forms. 2) . The largest amplitude of displacements was about 25 centimetres.2 Tests by German Railways: Results of a series of track buckling tests conducted for the Federal German Railways were reported by F. The buckling modes exhibited 2. Raab in 1960. buckling is the sudden lateral shift in the track alignment to release the built up compressive forces in the rail.50m and was confined at both ends by reinforced concrete blocks. Birmann and F. The force diagram after a buckle is shown in Fig 7.1) Buckling in the form of a ‘C’ could occur a on sharp curve (First wave form) while buckled track resembling an ‘S’ shaped curve is generally evidenced on straight tracks (2nd wave form). The studies conducted by various railways and the results thereof have been discussed in this chapter. The strength of track against buckling or what is described as lateral stability of track has been investigated in great detail by various railways. The test facility was located at the Technical University of Karlsruhe.2.(105) CHAPTER VII SPECIAL TOPICS 7. It indicates that while a track physically buckles over a length ‘l’ the force diagram is affected over a length ‘a’ where ‘a’ is several times ‘l’.1 Buckling Phenomena : As described earlier. 7.

7.1 .(106) First waveform Second waveform Third waveform Fig.

Research Department.3 Studies Conducted by British Transport Commission : In order to study the conditions and factors affecting the stability of the Long Welded Rails a large testing program was started in 1953 by the Civil Engineering Laboratory of the Western Region of British Railways. British Railways. the buckling load varied by as much as 25% Over a period of time with reversal of temperatures there is an accumulation of undesirable permanent lateral track deformations for temperature increases which do not cause actual track buckling but definitely increase the buckleproneness.L.2 .3. 7. Assistant Director of Research (Engineering). This is shown in Fig. 4) 7. 7.(107) Length of track where track has physically gone out of alignment 'l' Length over which force diagram has been affected 'a' 3) With use of different fasteners.Bartlett. These researches were carried out and described by Mr D. q Reduced compressive force Fig.

3 ACCUMULATION OF LATERAL DISPLACEMENTS . 7.(108) T4 & T1 are temperatures at which significant lateral displacements occur T5 & T2 are track buckling temperatures Fig.

4 120 ft length of track 2. Tie bars 6. 7. 9. Electric heaters 8.1 (109) 1. 7. Fig. Thermometers . End anchorage blocks Restraining beams 5. Hydraulic jacks Dial gauges registering longitudinal rail movement Dial gauges registering lateral rail movement 3. 4.

the whole capable of being subject to thermal stresses.4) The main tests devised for the purpose of carrying out buckling tests was a 120 feet test bed upon which could be built. although it must be stressed that the jacks were not directly used to induce compression in the rails. . two on each side of and clear of the test track. The latter was controlled by four tie bars. the longitudinal load required to buckle the track was determined experimentally for different types of sleepers. 7. The 120 feet track rails were anchored at each end to concrete blocks sunk below ground level.(110) 7. The length of misalignment is the length over which misalignment occurs. The arrangement of the test bed was such as to simulate the central portion of a long welded rail length on site which does not move longitudinally with temperature change. Methodology of Test Using the above setup. Misalignment : This is the offset of the rail from the straight. The track was laid initially as straight as possible and then given a small misalignment over a given length. they were situated on one side of each rail at a distance determined experimentally so that the rate of heating was not excessive. Four dial gauges altached to an independent datum registered any longitudinal movement of each rail end during the tests.1 Test Arrangement (Fig. This was sufficient to prevent rotation of the track and change of gauge but not to prevent the expansion of the rails. By operating the jacks the rail lengths could be kept sufficiently close to their original values to be consistent with actual conditions in the field. complete in every respect a length of track.3. Any tendency for the rails to expand could be counteracted by the jacks. Heaters : Electric heaters with parabolic refiectors were used to simulate the heat radiation from the sun. The test bed was laid inside a disused tunnel where a constant ambient temperature could be expected. Thermometer : Normal glass and mercury thermometers inserted in sockets drilled mainly in the head of the rail were used to measure the temperature.

3. Little can be done to this term. Using theoretical methods the longitudinal load required to buckle a track was determined and the same compared with experimental values. as it is dependent mainly on the properties of the rail. Where T is the torque resisting buckling and α is the angle of twist for the fastening due to rotation of the rail on the rail seat D is sleeper spacing q is the misalignment of the track over length L if Wmax is the lateral ballast resistance per meter length of track and W is the lateral ballast resistance per sleeper then Wmax =W/D Analysing the above expression.2 Buckling Load Formula: The formula derived for the longitudinal load required to buckle a straight track is : P= Π 2 EIs Π 2C + L2 16 D ΠL W max L2 + q Π 2q Where Is is the moment of inertia of the two rails put together in the horizontal plane. C is the torsional coefficient for the given type of fastening T = C α . Π 2C 2) 16 D ΠL represents the contribution of the sleeper/fastening q combination to the resistance against buckling. it can be seen that : 1) Π 2WIs represents the contribution of the rails to resistance L2 against buckling. 7. L is the distance between the points of contraflexure of the buckled track. Here clearly a reduction in sleeper spacing D or an increase in the fastening torsional co-effecient C will cause an increase in the overall resistance to buckling .(111) fastenings and ballast packing conditions.

2. 4.(112) W max L2 3) reperesents the contribution of the lateral ballast Π 2q resistance. Under normal conditions the percentage contributions could be 10%. In actual fact for a given combination of C. rail sleeper fastenings. only one thing resists such a rotational movement and this is the torsional resistance (denoted by torsional co-efficient C) afforded by the fastenings. 3. For practical use however. The buckling load values as determined experimertally show a fair correspondance (within a few per cent) with the values determined from theoretical calculations. and ballast would depend upon the actual conditions prevailing at site. in practice no track exists under these ideal conditions and a misalignment of ‘q’ over a length ‘L’ will always be present. However. the sleepers remain at right angles to the original track alignment. For this to occur. ‘L’ is chosen as 20 feet and the value of ‘q’ as 1/4 inch(6mm). Hence for various combinations of these variables. The following points are to be noted: 1. Wmax and q there exists only one value of ‘L’ which will yield a minimum value of ‘P’ (the buckling load). Clearly. the rail must rotate on the rail seat. Clearly the buckling load is proportional to torsional resistance. Experimentally it has been observed that when a buckling occurs. then the track would not buckle however great the longitudinal compressive force.D.30% and 60% respectively. If the track were perfectly straight and points of equal load application central for each rail. a range of ‘L’ values would emerge. ‘L’ the length of buckled track is taken as 20 feet for all cases. It means that large misalignments significantly reduce the strength against buckling. The relative contributions of rails. the smaller will be the buckling load. it is evident that the lower the L /q ratio. 5. In any case. .

The wave so created as seen in the vertical profile of the rail in front of the engine is called the precession wave. The effects of a moving train which could contribute to dynamic buckling are as given below : 1) Loaded axles of a moving train cause the track to be lifted in front of.4 Static Buckling and Dynamic Buckling The discussion so far has been centred on buckling caused by longitudinal compressive force buildup due to rise of temperature above the stress-free temperature. Reducing the lateral misalignment in the track. Providing full complement of ballast in the track as per precribed ballast profile.in the rear the recession wave and in between the axles. 2. Any of these waves Fig. in the rear of or even between the moving axles.(113) 6. Ensuring that no sleeper rail fastenings are missing.5 (a) . 3. Such a buckling is called dynamic buckling. the central wave. This buckling due to thermal loads alone is called static buckling. 7. The industry today is more concerned with buckling caused by the movement of a train on the track in the presence of thermal loads.A PWI can ensure that the track remains safe against buckling by: 1. 7.

7.5 (b) .DIRECTION OF TRAVEL q RECESSION WAVE PRECESSION WAVE q CENTRAL WAVE CENTRAL WAVE INITIAL LATERAL IMPERFECTION (114) ASSUMED LATERAL BUCKLING MODE Fig.

The model is depicted in the figure (Fig7.6 DYNAMIC TRACK BUCKLING MODEL . 7. 7.6).(115) could be critical enough to cause loss of contact between the ballast and the sleeper soffit resuting in the loss of lateral ballast resistance thereby making the track buckleprone (Fig7. This model is essentially a relationship between the lateral track displacement and the temperature increase over the force free or neutral temperature. 2) Tractive and braking forces applied by the moving train change the force level in the LWR and continuous braking at a given location could result in buildup of compressive forces creating buckling tendencies in the rail. 3) The hunting motion of the moving train over lateral mislignments in the track could create large lateral forces producing buckling tendencies. The model has 3 limbs as shown: AB is the prebuckling limb while D TBMAX Temp Rise above TN B 2 1 TBMIN C 3 TN A Displacement Fig. 4) Vibrations induced by the moving train could disturb the ballast and lower the lateral ballast resistance.5(a)&(b)).5 Dynamic Track Buckling Model: The effect of a moving train increasing tendency of a track to buckle when the temperatures are rising or the response of the track to disturbing lateral forces is depicted by what is called DYNAMIC TRACK BUCKLING MODEL.

and necessitates hot weather patrolling. Between TBMAX and TBMIN. the energy required to buckle a track is almost zero while below TBMIN the energy required to buckle a track is much larger than that which could be provided by a moving train. the allowable temperature increase TALL over the neutral temperature should be greater than the . Between points ‘B’ and ‘C’ a moving train could impart sufficient force to buckle the track. In the USA. Other software programs are CWRSAFE and related programs. At TBMAX. These could include : 1) Allowable Temperature rise above tn for maintenance activities 2) Temperature at which track enters the danger zone. It is assumed that if the track can be brought into position 2 it will automatically move into position 3. Using these programs the TBMIN and TBMAX of the given track for a given set of parameters are determimed. even a large force will not be able to buckle the track. the transition from the pre-buckling stage to the unstable buckled state and to the stable buckled state could be effected under the influence of energy imparted by the moving vehicle. For safe operations of CWR track with respect to buckling. At B when the temp.(116) BC and CD are post-buckling limbs. rises to TBMAX’ the track becomes unstable where even an infinitesimal lateral force will cause the track to buckle.6 CWR Safety Assurance Program: The discussion above could form the basis of a continuous welded track buckling safety assurance program. Between ‘B’ and 'C' the track on buckling will first move to an unstable buckled phase on curve BC and subsequently to a stable buckled phase on CD. Various softwares have been developed to predict the TBMAX and TBMIN temperatures for given track and rolling stock parameters. the program developed is called CWR-BUCKLE. The inputs to these programs are : 1) Rail section 2) Track curvature 3) Rolling stock characteristics 4) Lateral ballast resistance 5) The misalignment in the track. Below temp TBMIN at point C. The policies regarding LWR maintenance can then be decided. 7.

(117) difference between the maximum anticipated rail temperature TMAX on a given day and the neutral or stress-free temperature Tn i.e. TALL > (TMAX-Tn) The expression on the right is the anticipated rail temperature rise over Tn while TALL is the allowable temperature rise which could be determined from the values of TBMIN and TBMAX.. TALL will be somewhere between these two extreme temperature values depending upon the track parameters,level of maintenance and monitoring and the degree of risk the railway administration is willing to take. A conservative approach would be to fix the TALL at TBMIN value. However, a better approach would be to fix TALL higher than TBMIN if the railway has good track maintenance and monitoring procedures in place. The expression given above indicates that for a safe CWR assurance program two temperatures need to be determined: (i) TALL which is the allowable temperature rise above the neutral temperature for a given set of track and vehicle parameters.The single most significant factor which will govern TALL for a given set of track and rolling stock parameters is the lateral ballast resistance. The ralationship between TALL and the lateral ballast resistance will be in the form of a graph. This could be given to the field maintenance engineer to enable him to predict the allowable temperature rise over the neutral temperature for a given value of the lateral ballast resistance. (Fig 7.7) (ii) The neutral temperature or the stress free temperature of the track. 7.7 Field Determination of the lateral ballast resistance: A convenient metod to determine the lateral ballast resistance per sleeper has been developed in the USA.It is called the single tie push test (STPT). Test Methodology: The rail is freed from the sleeper by removing the fastenings and using a hydraulic jacking equipment the tie is pushed transversely to the track. With load transducers and gauges, the loads and corresponding displacements are recorded. The plot gives the maximum lateral resistance of ballast. For getting the average value the test could be performed on 3 sleepers over a 50 feet length. Once the lateral ballast resistance value is obtained

(118)

Fig. 7.7 Relationship between TALL and lateral ballast resistance. graph could be given to the field maintenance engineer to enable him to predict the allowable temperature rise TALL over the neutral temperature. 7.8 Neutral Temperature, Its Variation and Determination 7.8.1 Introduction : The neutral temprature tn or the stress-free temperature tois the rail temperature at which the rail is free of longitudinal stress, or longitudinal stress is zero. Till now the assumption was that once an LWR was destressed at temperature td, , td, was the stress free temperature of the rail. However, there is experimental evidence to indicate that the above assumption is not correct and the stressfree temperature of the rail tends to shift away from the destressing temperature. Accurate determination of the rail stress free temperature is of vital importance, because it is this temperature which determines the force level in the LWR

(119)

P = AEα [t P − t n ]
The above expression also indicates that if due to any reason the value of tn were to fall, it would automatically increase the compressive force in the rail and beyond a certain level could cause the track to buckle. Another way of putting it is that a change in the rail neutral temperature is equivalent to changing the force level in the rails for the same values of tP . 7.8.2 Factors which could cause a shift in the Rail Neutral Temperature: 1) Movement of the rail in the longitudinal, lateral and vertical directions : If the CWR were to be fully constrained, then there would be no change in the neutral temperature. Since the rails cannot be fully constrained in all directions, elongation or contraction can occur whenever the track is subject to train and environmentally induced loads. Railway track motions relevant to tn variation occur in the following three basic kinematic modes : (i) Rail longitudinal movement (ii) Track lateral shift (iii) Track vertical settlement. Consider a CWR being laid at Temperature tL and there is no rail longitudinal force at this temperature. Assume that the rail displacements (u,v and w in the longitudinal, lateral and vertical directions respectively) are measured with respect to an initial equilibrium configuration when the rail temperature is tL. These displacements may be due to a number of causes, and in many cases are not recoverable due to the inelastic nature of the ballast. From the displacements, the longitudional strain εx in the rail at any given temperature tP can be calculated from the fundamental equations of theoretical mechanics.,

 ∂u 1  ∂v  2 1  ∂ϖ  2  εx = −  +   +    + α [t P − t L ]  ∂x 2  ∂x  2  ∂x    

α [t P − t L ]

is a compressive strain taken as positive in the

longitudinal ‘x’ direction. The force in the CWR at tP in the longitudinal ‘x’ direction will be

by wheels negotiating a lateral imperfection or in case of curves. and are all tensile strains. then the neutral temperature will increase beyond tL. y and z directions may be due to the following reasons : i) Rail Longitudinal Movement – This may occur due to train action (braking and accleration) or due to wheel rolling action. . Track Lateral Shift – This may occur due to bogie hunting action. v and w the rail displacements cause compressive strains. ∂x ∂x ∂x If the movement of the rail leads to additional tensile stress. On the other hand. if u. they will cause tn to drop below tL . by vehicles operating ii) . On a curve of radius ‘R’ if the track is shifted by an amount equal to ‘V’ 1 then t n = t L + α  ∂u  V 1  ∂v  2 1  ∂ϖ  2        + +   +   ∂x  R 2  ∂x  2  ∂x     Movement of the rail in the x.(120) P=AE εx 2 2  1  ∂u 1  ∂v  1  ∂ ϖ      = AE α t P − t L +  +   +     -------1 α  ∂x 2  ∂x  2  ∂x         If tn is the neutral temperature of the CWR then P = AEα [t P − t n ] -------------2 Comparing equations 1 and 2 it is quite evident that 1 tn = tL + α 2 2    ∂u  1  ∂v  1  ∂ϖ       +   +   ∂x  2  ∂x  2  ∂x     where ∂ϖ ∂u ∂v .

3.1 A satisfactory neutral temperature measuring device should satisfy the following fundamental requirements : 1.8. the shift can be significant. . removal or trackline repairs all temperature. For tangent track. 7. 2. iii) Track vertical Settlement – Vertical wheel loads can induce differential settlement of the ballast which would cause development of longitudinal strains in CWR specially for new or recently surfaced tracks. the effect of lateral movement on the neutral temperature shift is likely to be small. Rail/Track Maintenance – activities involving lining. The rail residual stresses will affect the neutral temperature of the rail (Fig 7. will cause a shift in neutral Track maintenance lifting. application of rail anchors. The instrument should give absolute values and not relative values. plastic deformation occurs in the top layers of the rail head. Apart from these displacements which could cause a shift in the rail neutral temperature. however. The measuring instrument should be portable and not permanently attached to the rails. For curved track.8).3 Neutral Temperature Measurement 7. The basic mechanism involved is that the rolling contact loads change the residual tensile stresses in the top layer of new rails into compressive stresses. These are : . Rail “Rolling Out” – Due to vehicle loads. Experiments conducted by the British Rail showed rolling out of rail is pronounced in first 3 months after laying new rails and continues for about an year. Site specific calibration should not be involved. 2. 1.(121) in excess of or below the balance speed.8. two more factors could cause a shift.

any technique which relies on measurement of local stresses for the longitudinal force can have large errors.8 Rail residual stresses . The residual stresses are not associated with the rail longitudinal force since they are self-equilibrating in the sense that their resultant force and moment are zero. 5. fairly accurate with Fig. The technique should be non-destructive. The technique should be measurements within +10C.(122) 3. As a result. 4. The technique should be independent of longitudinal residual stresses in rail. 7.

Clearly a compressive longitudinal load will increase its deflection. Techniques under Research – 1. Accousto-elastic 4. Flexural wave propagation 2.3.(123) 7. 6.3. the following can be considered as reasonably developed : l Berry Gauge – Simple mechanical gauge to measure change in length. . Electromagnetic Accoustic Transducers 7. X-ray defraction 3. Magnetic coercion 5. It is based upon the fact that if the rail can be held at two points at some distance apart and a concentrated load applied at the centre of this portion. Barkhausen Noise – This principle is being used in Rail Scan Equipment. British Rail Vibrating Wire – Measures the rail force as a function of the frequency of a wire vibrating in a hole in the rail web.8. Wheatstone l ·l Strain Gauge –Uses a four arm bridge to measure the rail strains. whereas a tensile load will reduce it. the rail behaves like a beam column and its deflection is influenced measurably by the longitudinal load in the rail.3 Rail Uplift Method : A new approach based on rail beam column response has shown considerable promise. Laser ‘Spackle’ 7.2 Of the number of techniques available for neutral temperature measurement.8.

8.4 VERSE METHOD Practical use of this principle has been made in the technique called ‘VERSE’ developed by VORTOK International. EI. The first factor in the above equation represents the deflection under the concentrated load in the absence of any longitudinal rail force. As far as the end conditions are concerned. This is depicted in the given figures (Fig 7. Q and ∆ are proportional to each other. they depend upon the nature of constraint provided by the rig. UK and AEA Technology Rail. the end conditions are sufficiently repeatable. The second factor is the magnification factor due to longitudinal force. applied load Q and the nature of the end constraints.9 & Fig 7. Pc = Critical buckling load for the beam column of length 2L for the specific end conditions. Repeatability of the end conditions is an important consideration for successful application of the technique. the deflection is dependent on the rail flexural rigidity.3. Fixed support conditions improve the sensitivity. The equipment comprises of a frame featuring a hydraulic lifting device. The deflection ∆ is given by QL3 1 ∆=λ − − − − − − − (1) EI 1 − C Pc where C is the longitudinal compressive force in the rail Q = Verticle load applied at centre of rail.10) 7. the conditions are elastic supports (in between pure simple supports and completely fixed supports). It is necessary to design a rig such that for all locations and measurements. The above equation shows that for a given value of C (rail longitudinal force).(124) Besides the longitudinal force. a load transducer and a displacement λ . Generally. = Numerical constant value depending on the end conditions. but need large applied loads.

7.RAIL UPLIFT DRIVE (RUD) POWER PACK RECORDER AND PLOTTER (125) Fig. .9 Rail Uplift Method.

The measured data along with some other data such as ambient rail temperature. Taking measurements requires around 30 m of rail to be unclipped and placing rail support spacers at 10 m on either side of the measuring point (Fig 7.11).(126) T = 100 TONS T = 75 TONS T = 50 TONS T = 25 TONS T= tensile force C= compressive force Q= centrally applied load D = deflection at centre T = C = 0 C = 25 TONS C = 50 TONS C = 75 TONS Q C = 100 TONS D Fig. transducer. The height of the rail is included to take account of the . A maximum force of one tonne is applied and the load and displacements measured by the transducers relayed to the handheld computer. The rail must be in tension at the time of measuring the stress free temperature (SFT). The measurement systems are connected to a rugged handheld computer. 7. rail profile and height of rail is fed into the computer to obtain the SFT result.10 Graphical Method for neutral temperature determination.

11 VERSE METHOD FOR NEUTRAL TEMPERATURE DETERMINATION. .(127) Fig. 7.

Validation of VERSE technique has been carried out by AEA Technology.(128) rail head wear and rail grinding which will naturally affect the stiffness of the rail. . one of Britain’s leading technology companies.

. 12. 11. G. Longitudinal Ballast Resistance of Different Types of Sleepers Broad Gauge and Meter Gauge. Kish. Jeong 2. Long Welded Rails on Girder Bridges: RDSO/C-166. 10. RDSO/C-146.(129) List of References: 1. Report of Committee on Welded Rails – Civil Design Directorate Report No. Track/bridge Interaction Recommendations for calculations Modern Permanent Way – By M. Lateral Ballast Resistance of Different Types of Sleepers. 9. 4. Srinivasan. Manual of Instructions on Long Welded Rails 1996. 7. LWR on Girder Bridges with Creep Resistant Fastenings: RDSO/C-170. UIC Code 774-3R. 5. RDSO/C-169. Samavedam and D.1. 6. LWR on Girder Bridges. 8. ‘‘ The Neutral Temperature Variation of Continuous Welded Rails’’ By: A. Study of Rail Temperature in India. RDSO/C-156. Experiments on the stability of Long Welded Rails – British Transport Commission London 1961. 3. RDSO/C148.

Thompson. ‘‘How UP Achieves Lateral Track Stability on CWR’’ By William C. .(130) 13. 14. ‘‘ A New Safety Philosophy for CWR ’’ By Coenraad Esveld.

Sign up to vote on this title
UsefulNot useful