INCREASING TUNNEL LOADING GAUGE WITHOUT LOWERING THE INVERT.

KEYWORDS: Loading gauge, tunnel invert, LR55 track, track loadings, foundation behaviour

ABSTRACT

With few exceptions railway tunnels in the UK were built to 19^thcentury loading gauges, before the advent of piggyback euro standard HGV trailers, super cube ISO maritime containers, or electrification. Justifying the capital cost for the reconstruction of tunnels to larger loading gauges is difficult, even where "free" money from the EU is available. The alternative of lowering the invert is difficult, especially where tubular tunnels are involved. In any case only limited enlargement can be achieved.

Some rail lines in the UK can carry 8ft 6in high ISO containers, and with low floor wagons, 9ft 6in high containers. Neither the next generation of 10ft 6in high containers already in wide international use, nor 4m high, 2.5m wide and 12m long road trailers piggyback can be accommodated, without tunnel loading gauge enlargement.

This paper describes the development work of the LR55 track system, which can provide at least 300mm more head room in existing tunnels, within the existing invert. The LR55 track system is based on highway structural design philosophy and has been subjected to a battery of tests, including 80 tonne axles and 3m diameter tube tunnel loading.

1.0 INTRODUCTION

Since George Stephenson built railways, the sleeper with rails fixed either directly or on a base plate, has been the principal method of transmitting train loads into the sub soil via an elastic ballast. Brunel's longitudinal sleepers only found favour for bridges. Higher axle loads and train speeds, and the desire for better track standards with reduced maintenance costs, has seen traditional sleeper tracks improved incrementally with heavier rails, stronger fastenings, bigger sleepers and deeper ballast. On open tracks these improvements have been able to cope. In tunnels the situation is different, since the depth of ballast is constrained by the level of the invert, and so therefore are axle loads and speeds which can be accommodated, witnin a given loading gauge. In double track tunnels extra loading gauge can be won by singling but then there is a loss of train capacity, unless a new parallel tunnel is built. This paper discusses a new track system which requires a shallow foundation depth, even for the heaviest axle loads and high train speeds, and therefore promises larger loading gauges within existing tunnel inverts.

2.0 DESCRIPTION OF LR55 TRACK SYSTEM

Traditional track systems use bottom supported rails, which then need strong fastenings to sleepers to prevent overturning from lateral wheel loads. With discrete support on sleepers, trains experience regular hard spots at the sleepers, which are a cause of the formation of short wave corrugations on rail heads. Because spacing between sleepers is an order of magnitude larger than the sleeper width, ballast must be considerably deeper than that required for continuously supported rails. Finally rails and sleepers have to be adjusted together for line and level.

2.1 LR55 track components

There are six components to the LR55 track system (Fig. 1):

1. compacted highway type base
2. prefabricated concrete support troughs
3. gauging bars between troughs
4. ballast outside and between troughs
5. low profile, top supported rail
6. elastomeric grout to bond and support the rail in troughs

Figure 1. LR55 track system components

2.2 Track base

There is worldwide experience and expertise in the design and installation of compacted highway bases. The bearing quality, elasticity and life of such bases is well understood. There are many highway contractors competent to lay quickly such bases. These bases can be laid with a very high tolerance of material quality, and line and level. This is important, as the pre cast concrete support troughs sit directly on the base. The level of base, together with the prefabrication tolerance of the troughs determines the first order accuracy of the ultimate line and level of the track. The troughs are gauged together by bars, although the mass of the ballast, the lateral stiffness of the troughs and large trough sides, provide substantial lateral restraint to the track.

2.3 Concrete Support Troughs

The support troughs are vertically and laterally stiff. This is important as the rail is less stiff vertically than girder rails (eg. UIC60). The wide base of the trough, and the continuous support, means that the trough pressure into the track base typically lies between 150 and 250 MPa, for 25 tonne axles loads, depending on the stiffness of the base. Even a weak base however, like sand has a load capacity of about 5000 MPa, an order of magnitude greater than the imposed pressure.

Support Troughs are manufactured and delivered to site in lengths to suit handling and installation to the required quality. At this stage it would seem sensible that support troughs would be 6m long, which weigh about 600kg. Longer Support Troughs could be tried as part of an optimisation exercise balancing import material costs against site handling and adjustment. On curves, support troughs are laid as tangents to the required radius and design super elevation for the train speeds expected. Simple mechanical links, fix the ends of support troughs together, until the application of a bonding grout permeates between the trough ends and bonds the troughs together.

There are various methods available for prefabricating the support troughs. Many manufacturers are capable of producing the troughs to the required quality of concrete and dimensional tolerance. Pre-stressed tendons primarily fulfil the function of ensuring that troughs can be delivered to site intact. Most of the track testing however has been undertaken satisfactorily with unreinforced troughs.

2.4 LR55 rails

The LR55 rail is top supported in the pre-cast concrete trough. This makes the rail very stable and highly resistant to overturning. About 60% of the wheel loads are transmitted on the running side rail flange, about 30% on the non running side flange, and 10% by the rail base. The rails weight about 55kg per linear metre and therefore have similar electrical resistivity to girder rails of similar weight. Lateral train loads are accommodated by shear compression of the elastomeric bonding grout, and the lateral stiffness of the troughs and track construction.

The LR55 rails are welded into long strings and pre tensioned longitudinally to compensate for ambient temperature variations. The rail welds are located away from joints between the support troughs, to prevent hinges being created. The rails are supported temporarily either by stands, or wedges of pre-cured bonding material. The rails are adjusted for line, level and gauge.

Once the rails are to line, level and gauge, they are bonded into the concrete support troughs by elastomeric grout. This is the second order determinant of track accuracy. It should be possible to achieve a tolerance of 0.1mm. There is no mechanical connection between rail and trough, with the rail continuously supported vertically and restrained laterally by the elastomeric grout, rail/wheel interface forces are almost constant along the track. This should mean a better ride for vehicles (and their loads) as well as reducing the incidence of long, medium and short wave length corrugations along rail heads.

The LR55 rail is made with a built in continuous check rail. This should reduce derailments, especially by flange climbing over the rail head, since the other wheel on the axle will be restrained laterally by the check rail. In the event of a derailment, trains cannot drop because the track formation is level with the top of the rail. This also means that derailed trains are unlikely to overturn off the track.

Finally on curved tracks with high speed running, gauge corner cracking is less likely to occur, since the centrifugal wheel forces are shared between outside and inside wheels on the curve. In the less likely case of cracks progressing to failure, as occurred in October 2000, the bonding and support trough which surrounds the rail will prevent it falling apart, and thus denying the derailment mechanism which had tragic consequences at Hatfield. This continuous support for the rail, means that the more common weld failure need not be so critical, since the broken rail ends will be restrained and kept together.

2.5 Elastomeric Grout

The rails are bonded into the support troughs with an elastomeric grout which transmits the static and dynamic forces from wheels through the trough into the rail base. Polyurethane elastomers are now widely used in rail application, eg. noise reducing base plates. There are many proprietary polyurethane bonding grouts available "off the shelf", fully tested for the climatic and loading conditions experienced in railway environments.

3.0 COMPLETED MODELLING

3.1 Modelling

Static and dynamic load models were examined. Wheel loads rolling along and across the rail have been modelled. The outputs of the models are:

• rail deflection
• trough deflection
• shear forces in rail
• shear forces in trough
• bending moments in rail
• bending moments in trough
• pressure at base of trough

3.2 Analytical Models

Modelling conventional railways has been undertaken by considering the rails as a single layer beam, and the sleepers and ballast as a homogeneous elastic foundation. In the LR55 track system the rail is continuously supported in a concrete support trough. This is better represented as multi layer beams with elastic foundations. Here the rail and the support trough are considered to be beams, and the elastomeric grout and sub-base as elastic foundations with different moduli. Analytical models can only solve a limited number of idealised problems. The boundary conditions of these were discussed by Hetenyi (1946). Timoshenko et al (1932) had earlier analysed the stresses in railway tracks in the same way.

3.3 Finite Element Method

The use of the Finite Element Method (FEM) allows a wider range of problems to be solved, with different loading and boundary conditions, and non-linear foundation properties. These were applied to railway problems by Fateen (1972) and explored by Miranda et al (1996).

The analysis of the LR55 track system using FEM is based on a stiffness approach for solutions, the nodal displacements are assumed to be the basic unknowns. The nodal equilibrium may be expressed by the stiffness matrix equation (1):

Solving this equation with FEM requires the track to be divided into a number of elements. The contribution of the track base to shear resistance in the stiffness matrix is so small and unreliable that it can be ignored. The stiffness matrix of the LR55 track is therefore the assembly of the stiffness matrices of all its components.

3.3.1 Rail and concrete beam elements

The steel rail and concrete support trough can be treated as conventional beam elements, with two nodes per element. Each node has three degrees of freedom; horizontal displacement (u), vertical displacement (v) and rotation about the z-axis (ø). This produces a stiffness matrix, which is therefore 6 x 6. Przemieniecki (1968) discussed the coefficients of the stiffness matrix.

3.3.2 Pad Spring Elements

The elastomeric grout acts like a pad and is represented by a number of discrete vertical and horizontal springs. Each spring has one degree of freedom per node which is displaced in the axial direction. The vertical springs are assumed to be of a Winkler type, Selvadurai (1979). These can be defined as equ. (2)

The two ends of the grout vertical spring elements are free to displace, so the stiffness matrix is 2 x 2 in the form :

			vi	vj
	[K]e = K1v		1	-1		vi		(3)
			-1	1		vj

3.3.3 Track base vertical spring elements

The track base foundation can also be considered to consist of Winkler type vertical springs. Each spring having one degree of freedom. The stiffness of each spring is then equ. 4:

3.3.4 Input assumptions

These equations were used to predict the behaviour of the LR55 track specified with the following characteristics:

The track base modulus is assumed to be 20 N/mm² and the Young's modulus of the elastomeric grout is assumed to be 2.42 N/mm². The grout is assumed to be uniformly 20mm thick. The applied wheel load is 122.6 kN, equivalent to the 25 tonne maximum axle load permitted on Britain's railways.

3.3.5 Deflection

The maximum deflection of the LR55 rail under the above conditions is 7.8mm, and occurs at the point of wheel loading. The deflection decays to zero over a distance of 2m from the point of load. There is then a negative deflection (hogging) where the rail rises above the neutral position by up to 0.1mm over a further 2m length.

Similar characteristics can be seen in the behaviour of the concrete support trough, although the deflection is much less. Immediately under the wheel load deflection is at a maximum, of only 3mm. Again there is a hogging of 0.1mm between two and four metres from the load. These two calculations show that at the point of load the rail is depressed into the grout and trough by about 5mm. This is below the 6mm figure specified for American main line railways . (Fig. 2)

3.3.6 Bending moment

The maximum bending moment in the LR55 rail occurs, as would be expected, under the wheel load. Here it has a value of about 20kN.m and then declines away from the load to zero 0.5m away. It continues to decrease to a negative moment of about 4kN.m , a metre from the wheel load. The moment increases and approaches the neutral axis smoothly regaining neutrality about 3m from the load.

For the concrete support trough there is a similar picture, except that the bending moments are much less. Under the wheel load it is 7kN.m, and the maximum hogging of 2kN.m about 1.5m from the load. However in the case of a FEM for a trough only 4m long, the end effects of the more rigid trough show that hogging continues to decrease to a value of -6kN.m. (Fig. 3 )

Figure 2. Deflection of LR55 rail and support trough under 25 tonne axle load

Figure 3. Bending moment of LR55 rail and support trough under 25 tonne axle load

In all cases, the FEM and analytical model results closely agree. Given the complexity of the FEM, for first order approximations the analytical beam model gives very satisfactory results.

3.4 Noise and vibration transmission modelling.

A mass-sping-mass-spring model for noise and vibration attenuation from wheel/rail interaction into the ground has been constructed. This predicts a 30dB reduction in the range 0 - 20 Hz, and 50dB reduction in the >100Hz range compared to girder rails on sleepers in ballast. This is due to two main factors. There is no rail web to act as a sounding board. Secondly the mass-spring-mass-spring transmission path provides the optimum attenuation, with noise and vibration energy absorbed by warming of the elastomeric grout.

3.5 Thermal Forces

The effects of temperature variations in the LR55 rail, elastomeric grout and concrete support trough was modelled by differences in expansion (and contraction), on the equilibrium in the track. These exert shear forces along the LR55 track, between rails and bond, and bond and trough. Further forces act in extension (or compression) between the rail and bond, and bond and trough. For the temperature range examined neither of these family of forces will result in bond failure.

3.6 Laboratory Tests

A series of physical tests were undertaken in the Civil Engineering Structures Laboratory of the Liverpool JM University. The tests used a standard hydraulic press able to deliver up to a 500kN (about 50tonne) load, either statically or dynamically. Track samples 1m and 6m long were tested with static and dynamic loadings. In all cases the tests pieces were instrumented with strain and displacement gauges. The applied loads were measured with a load cell. All data were recorded with a computer, which was used to analyse the results obtained. Tracks were also tested on a variety of base materials and a range of ambient temperatures from -10°C to +60°C. Half of the tests were conducted with the tracks under water to check for elastomeric bond delamination and track failure by pumping. Two tests to destruction were undertaken, to examine the behaviour of the track in failure mode.

3.6.1 Description of Rail Samples Tested

Sample rails, 1 metre long were cast by Edgar Allen Engineering Ltd. from standard manganese steel used for normal rails, points, switches and crossings. The rails had three different depths 60mm, 70mm and 80mm. The first and third samples had plain rounded bases (as in Fig.1), while the second had notches in the rail side (Fig.4). A further cast rail 80mm high and 6m long, with a rounded base was made for foundation failure testing.

Figure 4. LR55 rail profile with notched sides

The 1m long rails were bonded into a weak mix unreinforced cast concrete support troughs using either rubber filled polyurethane grouting (KC330) provided by SIKA Ltd, or System 6 polyurethane grouting provided by ALH Ltd. Attached to the underside of the rails were strain gauges which were placed 0.2, 0.5 and 0.7 metres along the length of the rail, in the centre line.

Some of the tests were conducted in dry conditions. In others the rail and concrete base were covered by 25mm of water retained by a tank created around the concrete base and sealed with silicon mastic.

The 6m long sample was comprehensively strain gauged along the length of the rail and the support trough, on the top, sides and bottom.

3.6.2 Description of the Test Rig

The 1m rails embedded in their concrete bases were mounted on the hydraulic impulse press normally used to test highway pavement specimens. A load cell was placed over the centre of the upper surface of the rail and loads imposed which could be raised in 0.1 kN steps. The load into the rail head was imposed using a large steel ball to replicate a rail wheel. Imposed loads were both static and dynamic. The strain gauges plus a displacement gauge at the centre of the rail were connected to a data logging system.

3.6.3 The Endurance Testing Procedure

At the start of each test, increasing static loads were imposed in incremental steps from 0.0 to 100 kN and reference readings from the displacement and strain gauges taken. The rail was then subjected to 1.25 million cycles of sine wave dynamic impulses with a frequency of 15Hz, and a maximum load of 50 kN.

A further graduated load test up to 100 kN was then undertaken. The rail was then subjected to another 1.25 million cycles. Lastly the graduated load was repeated. The before, middle and after results were then plotted for comparison, to determine what deterioration of the track had occurred. 15Hz was selected to represent an operating train speed of over 100 km/h.

This procedure was then repeated for a further 2.5 million cycles with the sample rail submerged in water. Giving 5 million cycles in total for the sample. The other 1m rail samples were subjected to the same procedures as the initial rail, each having 2.5 million cycles, dry and 2.5 million cycles wet. Giving an accumulated total of 15 million cycles for the initial three rail samples examined.

The 6m long rail sample bonded into a 6m long concrete support trough was placed on a prepared sub base over 1m deep, to replicate the rail sub soil. This sample was subjected to static and dynamic loads as for the 1m long samples. This sample was also tested with a 1m wide void in the sub base immediately under the point of load application.This was to determine the ability of the LR55 system to resist weak and collapsed foundations and sub bases. All tests on this sample were undertaken with 25 tonne axle loads.

3.6.4 Simulated loads

Simulated axle loads up to 80 tonnes were used on the LR55 track without failure. Cyclic loadings with 25tonne axle loads using a 15Hz frequency (simulating a passing speed of about 30m/sec.). In total all track samples were subjected to about 30million axle loads, without showing any signs of failure.

3.6.5 Static tests.

Static axle loads have been tested without failure up to 80 tonne, and dynamic axle loads of 25 tonnes. Track samples have been subjected to endurance tests of 30 million axles passing at simulated speeds of over 100km/hr. Samples have been tested under water to ensure no penetration, or failure from pumping or frost damage.

One track sample was tested in a temperature chamber, where the ambient temperature was varied from -10°C to +60°C. This allowed both the performance of the track to be examined under different temperatures but also the variation of stiffness to be examined, of the elastomeric pad between rail and concrete support trough.

All tests have been undertaken on 8 different track samples at up to the equivalent of 80 tonne axle loads. Two different PU grouts (SIKA KC330 and ALH System 6) have been tested for performance. Three different rail heights (60mm, 70mm and 80mm) were tested, as well as rails having notches in the lower rail side to determine what difference of performance resulted. After these tests, the 80mm rail section was selected for further testing, without side notches. The analytical structural models have been validated on a 6m long track sample.

3.6.6 Dynamic tests

Tests were undertaken at a simulated train operating speed of over 100km/hr with an equivalent of at least a 20 tonne axle load. This was achieved with a load of 100kN applied sinusoidally at a frequency of 15Hz. In total some 30 million passing axles have been simulated. About 10 million axle loads were tested with the track samples flooded, to determine the degree of water penetration and possible failure. No water penetration or failure was recorded.

One sample was tested dynamically in a temperature chamber in the range -10°C to +60°C. This recorded the hardening at lower temperatures and the softening at higher temperatures of the bonding PU grout. This did not impair the performance of the track system, which showed no signs of premature failure, having previously been tested under wet and dry conditions (3.2.3).

3.6.7 Test results

These results have shown that from laboratory tests using a hydraulic impulse press to simulate 126 kN loads at 15Hz, equivalent to over 100 km/h, sections of LR55 low profile rail with different rail depths were able to withstand and recover from the loads imposed, which were equivalent to maximum main line railway standards. A total of 30 million simulated axle loads were imposed. The behaviour of the PU bonding grout replicated earlier tests at the University of Calgary (Shrive & Ameny 1987). The PU grout was able to withstand the ingress of water from 2.5 million cycles submerged for each sample tested.

The rail and bonding material behaved according to Hooke's Law, and no evidence emerged that the elastic limit was approached for any of the samples. Nor was there any evidence of a deterioration in the performance of the PU grout, which continued to behave elastically. However, these laboratory tests did not simulate a rolling axle passing along the rail or the impact of sine wave deformation and harmonics created by adjacent axles.

3.6.8 Failure tests.

Destructive tests were also undertaken. A one metre sample was supported at each end as a simple beam. It was loaded in its centre to failure. The load was increased in 10 kN increments. The track failed at 296 kN, a simulated axle load of about 58 tonne . The concrete of the trough failed in tension. The bond between the polyurethane grout and both the LR55 rail and concrete support trough did not fail.

A second destructive test was of the strength of the PU grout to resist the rail pull out from the trough. This also determined the resistance of the PU grout to thermal expansion forces of the rail to high ambient temperatures, which could cause the rail to be displaced from the foundation support trough. The test involved pulling out, with an incrementally increasing load, a 1m long rail from its trough. The one metre long sample failed at a pull out load of 29kN (about 3 tonnes). Again the concrete in the trough failed in tension. The elastomeric bond held. This means that the LR55 can resist thermal expansion from atmospheric conditions experienced in a 70°C temperature range .

In both cases the elastomeric grout did not fail. The grout remained bonded to the LR55 rail and to the concrete trough. The concrete troughs failed in tension. Both tests indicate considerable over capacity compared to loads to be met in normal railway operation.

3.6.9 Electrical Resistivity

The control of stray currents in normal conditions is determined by the electrical resistivity of tracks. One metre long track samples in the laboratory measured better than 20,000 mega Ω resistance. The resistivity of both PU grouts tested were similar. This translates in the field to over 10,000 Ωkm resistivity, and would suggest negligible stray currents in normal conditions.

3.7 Site Testing

Two field trials have been successfully mounted.

3.7.1 Rotherham Bus Station.

A 10 metre length of LR55 track was installed in March 1993 in the entrance to Rotherham Bus Station, were buses crossed, turned around and passed along the rail. The trial was concluded in Sept. 1995 after about 2 million buses had impacted on the LR55 trough. It was laid in three support troughs; 2.5m, 3.0m and 5.0m long to determine the behaviour of different trough lengths under load.

The rail and trough were instrumented with strain gauges and the results captured on a data logger. Regular analysis of this data showed that the track remained in good condition. After a year all gauges had exceeded their design life but continuing visual monitoring confirmed that the LR55 track had not deteriorated or failed. This trial represented about 30 years traffic in a typical urban radial road.

This was also confirmed by a trench dug under the site in Sept. 1994 to expose the track construction. The concrete trough showed no signs of failure or deterioration. This was inspected by interested parties including the Railway Inspectorate, public utility representatives etc. The trench showed the shallow construction of the track and that the LR55 track is self supporting over trenches (or foundation voids).

3.7.2 South Yorkshire Supertramway.

Over a weekend in March 1996 a length of LR55 track was installed on a single track section of the Supertramway where a conventional girder railed track had failed after only two years. The LR55 installation was at a point where some 300 LRVs and 100 HGVs impact per day ( a total of about 120,000 vehicles pa, or about 6million tonnes pa). A failure of this sample would halt the main line to Meadowhall. The track was installed one rail per night time possession, without interruption of day time rail service or road traffic. This installation has been monitored regularly, and the alignment measured for stability. The installation was accepted by South Yorkshire Supertram Ltd. into its maintenance responsibility in Sept. 1996. The LR55 track continues to give a satisfactory performance

3.7.3 Heavy Rail application

A presentation and discussions have been held with Deutsche Bahn for the use of the LR55 track system in one of two particular uses:

1. Low cost, low maintenance, low traffic branch lines
2. High Speed slab track, with troughs for easy rail replacement, and train containment in case of derailment.

The development and testing work have been undertaken to mainline railway standards. Even at 25 tonne axle loads, the pressure at the base of the foundation trough is low (<250 mPa), and by distributing loadings over a wide area, means that track should have a longer life than conventional sleepered track. Gauge maintenance is achieved by a combination of ballast resistanc, trough stiffness and gauge bars between the support troughs.

4.0 APPLICATION TO TUNNEL LOADING GAUGE ENHANCEMENT

The total height of the LR55 track system above the sub soil/sub base is only 200mm. This is at least 300mm lower than sleepered tracks. A lower track height means that a larger loading gauge is achievable in tunnels without the need to lower the invert or reconstruct the tunnel. (Fig. 5).

The LR55 track with a level track surface also provides other benefits for tunnels. These include:

1. safer staff working conditions
2. safer emergency train evacuation
3. ability to use road maintenance or rescue vehicles without modification

4.1 Tubular tunnels

The LR55 track benefits of increasing loading gauge are available for tubular tunnels, where the cost of lowering the invert would be considerable and disruptive. Indeed a section of London Underground tube tunnel (3 m diameter) was replicated in the Structures Laboratory at Liverpool JM University to investigate the stresses in the lower tunnel sections and the ground around the tunnel. (Fig. 6) As in the above tests, this showed that the LR55 track could create an increase in tunnel loading gauge height of over 300mm, without increasing the stress in the tunnel segments, or overloading the surrounding ground. A (temporary) transition section between LR55 and sleepered track has been designed to enable relaying with LR55 to be undertaken on a progressive basis with short possessions if necessary.

Figure 5. LR55 track in existing tunnel with enlarged loading gauge

Figure 6. LR55 track system in tube tunnel to enlarge loading gauge (figure to come)

4.2 Other tunnels

Masonry tunnel walls with strip foundations , or built onto bedrock represent a different structural challenge for increasing loading gauge. While lowering the invert is physically possible to achieve, the cost of disruption while the operating railway is closed or curtailed makes this also difficult to achieve. The advantage of the LR55 track is that it can be laid in place of existing tracks on the same possession basis, with immediate reopening of tracks, albeit at temporary lower speeds to accommodate the transition between LR55 and remaining sleepered tracks. The increased head room and loading gauge is particularly relevant to new rail freight traffics, eg. piggyback, and the new 10ft 6in high ISO containers. (Fig.7) now in use on maritime trades.

Figure 7. LR55 tracks accommodating larger loading gauge for freight trains (figure to come)

5.0 OTHER RAILWAY APPLICATIONS OF LR55 TRACK SYSTEM

The LR55 track system, while providing a specialist solution to increasing tunnel loading gauge without lowering the invert, does have other practical applications for main line railways, some of which are topical.

5.1 Gauge corner cracking and other rail failures

This is a well known phenomena (eg. Profillidis 2000, p.104), and different railway administrations have adapted a variety of treatments to prevent rail failures through microscopic cracks progressing to catastrophic failure. Rails fail for other reasons, eg. weld failure, fishplate failure. In 1960 rail breaks in Britain averaged 200 per annum (Hall 1992). However within ten years the figure had increased to over 400 breakages pa., in spite of a smaller network. In addition there were over 200 weld breaks on CWR. By the late 1980's the rail breaks had decreased to 350pa while the CWR weld failures had increased to over 250 pa. Inspite of ultra sonic testing, microscopic cracking can occasionally progress to complete failure as at Hatfield (Oct. 2000) with tragic consequences. Unfortunately a rail failure is one of the few examples in railway technology which is not "fail safe". If a rail physically fails, trains derail.

In the LR55 track system, the rails are completely retained in the elastomeric grout, which is itself contained by a substantial and stiff concrete trough. If a rail (eg. weld) should break, the rail ends will be restrained. This is due to the continuous support of rails in the LR55 system. So the location of a random rail failure will be immaterial, the rail ends will remain nearly to line and level, and thus passing rail wheels should stay on the rail. Further the train wheels on the other rail are also restrained by the continuous check rail, so that if the failed rail is being pushed outwards , the wheels (and train) cannot follow.

Should a rail failure lead to a derailment in the LR55 system, the rail vehicle will arrive on the ballast which is a rail height. Trains are therefore unlikely to fall over, which is another cause of casualties in many rail crashes.

5.2 High speed tracks

High speed tracks using heavier sleepers and rails, together with deeper ballast require considerable and regular maintenance to provide an acceptable line and level for a comfortable ride for passengers. There is a trade off between low first and high maintenance costs, and high first and low maintenance costs. In an attempt to make the latter more economically attractive, a study has been undertaken with the LR55 track as part of a slab track system (Fig. 8).

Figure 8. LR55 track in slab foundation for high speed trains (figure to come)

Continuous reinforced concrete slabs have long been in use for highway applications, and specialist machines, eg. slip form paviour, mechanise the production of the slab to a high tolerance of line and level. The slab is structurally more stable than sleepered rails, since the slab is very stiff in the lateral plane, across the track, which is most sensitive to movements out of alignment. Indeed this technique as a continuous extrusion of concrete can be modified to provide any profile. This was used in Sheffield to provide the track bed for the new light rail system. The slab must however be correctly laid, since this determine the line of the rails.

The advantage of marrying the LR55 track system to a continuous slab can be summarised as follows:

1. high tolerance for rail line, level and gauge
2. reduced chances of derailment due to continuous check rail
3. in derailment train does not fall off track,
4. minimal slab damage in the case of derailment and thus
5. no needed to replace slab after (train) accident
6. ease of rail replacement after accident.

In the light of the Hatfield crash, slip formed slabs for curves combined with LR55 would reduce maintenance costs and the risks of derailment.

5.3 Low maintenance cost branch lines

For branch lines the dynamic challenges on the track are rather less than on high speed. Nevertheless the track must be able to accommodate 25 tonne axles at low speed. For branch lines low first cost and low maintenance costs must combine to improve the overall economics in the face of lower traffic volumes. This could be particularly important where closed line restoration is being considered. Here a variation of the LR55 principle has been developed exploiting highway construction technology (Fig. 9).

Figure 9. LR55 tracks for branch lines (figure to come)

The starting point will be the laying of a standard road base, the thickness determined by the CBR of the subsoil, the annual train tonnage and the required track base life. The compacted road base will have a high bearing strength together with elastic structural properties. The base can be laid to a high line and level tolerance. On this base, sit the LR55 support troughs directly. The troughs are joined by gauge bars to provide track gauge. The LR55 track on road bases have been tested without failure up to static axle loads of 80 tonnes. The LR55 troughs are stiff in the vartical and horizontal plane and across the track. To provide further resistance to the track going out of alignment, ballast is laid outside and between the troughs to anchor the track on its base. Finally the rails are bonded into the support troughs and a very high level of line, level and gauge tolerance can be achieved at this stage. This improves ride stability and therefore reduces of the mechanisms for track being pushed out of alignment, rail damage or premature failure.

5.4 Level crossings

There will be railway level crossings for the foreseeable future. Level crossings are the responsibility of the rail infrastructure owner. Level crossings are a perennial cause of problems and complaints, especially for road traffic crossing the rail line. This arises as road pavements are not tied into rail tracks and therefore heavy road vehicles can cause the misalignment of both road and rail ways. The LR55 track addresses this directly by making the road pavement the foundation for the rail line (Fig. 10). A transition rail has been designed to accommodate the joining of LR55 rails to all current girder rail profiles.

Figure 10. LR55 track for Level Crossings (figure to come)

6.0 CONCLUSION

The LR55 track system provides a number of unique opportunities to improve the safety, infrastructure and economics of railways. These can be summarised as follows:

1. loading gauge enhancement in tunnels and under bridges without lowering inverts
2. high speed curves to reduce gauge corner cracking failures
3. high speed tracks to reduce costs and improve alignment
4. low cost branch line (re)tracking
5. low maintenance level crossings.

The LR55 track is as radical a change in rail track technology, just as the Vignoles rail was to bull head railed tracks in the 19th century. The short and long term economic advantages are probably greater, especially the opportunity created to attract freight traffics which cannot at present be carried by rail.

References

Al Nageim H.K. & Lesley L.
Investigating the structural reliability of a new low profile rail system under simulated environment.
Proceedings of 18th Energy Sources Conference, ASME International, Structural Dynamics and Vibration. Houston Texas USA Vol. 170. 71-77

Fateen S
A finite element analysis of full depth asphalt railway tracks.
MSc Thesis,
University of Maryland. 1972

Hall, S,
Railway Disaster - cause and effect. Vol. 2. p.195
PRC Ltd. 1992,
ISBN 1 85648 049 6

Hetenyi M
Beams on Elastic Foundation.
University of Michigan Press. 1946

Lesley L
Light rail transit developments in the UK.
Rail Engineering International, 1993 Vol. 3. 2-4

Lesley L & Al Nageim H.K.
Transmission of vibration and the conditions of resonance in the LR55 track system.
Proceedings of 18th Energy Sources Conference, ASME International, Structural Dynamics and Vibration. Houston Texas USA Vol. 170. 233 - 236

Miranda C & Nair K
Finite beams on elastic foundations.
Journal of Structural Division ASCE. 1996 Vol. 92 No. ST2 131-142

Mohammad F, Al Nageim H.K., Lesley L & Pountney D.
A theoretical analysis of LR55 track system as Multi layer Beams on elastic Foundations: An Analytical Approach.
Proceedings of International Symposium on Theories and Applications of Traffic and Transportation Systems Engineering. Beijing, PR China 12-13 July 1996. 113-126

Profilliddis, V.A., Railway Engineering
Ashgate, Aldershot, UK 2000
ISBN0 7546 1279 1

Przemieniecki J.S.
Theory of Matrix Structural Analysis.
McGraw-Hill. New York USA 1968

Selvadurai A.P.S.
Elastic Analysis of soil-foundation interaction.
Developments in Geotechnical Engineering.
Elsevier Science Publisher.1979

Shrive, N.G. & Ameny, P,
Tests of resilient track support materials for City of Calgary Light Rail Transit,
University of Calgary Aug. 1987

Timoshenko S & Langer B.F.
Stresses in Railroad Tracks.
Transactions of the American Society of Mechanical Engineering.
USA 1932 Vol. 54 277-302

Figures

1. plain LR55 track
2. LR55 track displacement
3. LR55 bending moment under load
4. notched LR55 track
5. tunnel loading gauge
6. tube tunnel cross section
7. piggy back in tunnels
8. high speed track
9. branch lines
10. level crossings