The present invention relates to a railway-rail induction welding device, in particular a railway-rail induction welding device for on-site mobile welding of railway rails, for example during repair or installation of railway rails. The invention further relates to a railway-rail induction welding apparatus and a method of joining two railway rails.

Welding of railway rails on-site or in situ, or in a static welding facility or factory may be required.

For example, defects can be present in railway rails and may need to be expelled, which may be achieved by welding and forging the section of rail.

Additionally, when installing railway rails or replacing a section of railway rail, it is necessary to join two railway rails together. It is desirable to weld the railways together, rather than fastening them together, since this provides a smoother running surface.

Furthermore, it may be desirable to shorten a section of continuously welded rail.

Railway rails are conventionally welded onsite using alumino thermic welding or electric resistance welding. However, alumino thermic welding can result in a poor-quality weld and electric resistance welding, such as flash butt welding, requires a good electrical contact to be effective, which can be challenging to maintain.

Additionally, alumino thermic welding can introduce non-parent material products, and flash butt welding can introduce inherent defects. Gas pressure welding and other welding techniques may also introduce defects. Heating and forging of railway rails may be required to treat these previously welded joints. This may be to expel the previously welded material. The heating and forging process may also be used to remove rail material which has a surface defect.

It is therefore desirable to inductively weld railway rails, which involves heating the railway rails via electromagnetic induction, in particular high frequency magnetic field induction, with heat being generated in the rail by in the rail by hysteresis losses and eddy currents. This provides a good-quality weld and does not require electrical contact with the rails.

An inductor, which may also be termed a coil, with an alternating electrical current passing therethrough is used to generate a magnetic field. The term “coil” does not necessarily require a specific coiled geometry, and instead may generally refer to an inductor.

Efficient and reliable inductive welding can be difficult to achieve. For example, ensuring uniform inductive heating of the portion of the rail can be challenging due to the irregular geometry of railway rails. Additionally, the inductive welding apparatus should be easily positionable around the rail ends to be joined or around the rail with a defect to be removed. The coil or inductor should be made from a material which is sufficiently conductive to permit the generation of a strong enough magnetic field to cause melting of the railway rail, whilst also being configured to resist melting or deformation from resistance heating or radiant heat from the railway rail.

The present invention seeks to provide a solution to these problems.

According to a first aspect of the present invention, there is provided a railway-rail induction welding device comprising: a copper tubular body having a cooling-fluid inlet and a cooling-fluid outlet; the copper tubular body having a rail-facing rail-heating portion and a cooling-fluid return portion; the rail-facing rail-heating portion having: a longitudinal extent which is non-linear and profiled to correspond or substantially correspond to one side of a railway rail to be treated, and a lateral extent which is non-uniform along the said longitudinal extent.

Copper has a suitable resistivity which allows for generation of a magnetic field sufficient to inductively heat the rail to a predetermined welding temperature. The tubular body allows for water cooling to prevent the copper melting due to resistance heating or radiant heat from the rail.

The longitudinal extent of the body will be understood to be defined by the direction which extends along or parallel to the flow path of fluid or electricity through the tubular body. This is effectively in a top-to-bottom direction along the side of the or each railway rail. Whilst the longitudinal extent is described to be profiled to correspond or substantially correspond to one side of the railway rail, it will be appreciated that this does not exclude a non-uniform separation between the railfacing rail-heating portion and the railway rail. In other words, the longitudinal extent does not need to exactly correspond.

By corresponding to one side of the rail, two devices can be conveniently positioned either side of the rail which allows for straightforward removal after welding.

The lateral extent will be understood to be the direction in an axial or longitudinal direction of the rail. The lateral extent of the rail-facing rail-heating portion being non-uniform allows for portions of the tubular body to be wider than other portions. The local magnetic field density is reduced at the wider portions compared to the narrower portions. The locally reduced magnetic field density results in locally reduced induction heating in the rail. This is advantageous for inductively heating a railway rail since, for example, railway rails have an irregular geometry and thinner portions of the rail may require less heating to reach a predetermined welding temperature than thicker portions. The wider portions of the tubular body can be arranged to correspond to positions where less inductive heating is required to reach the welding temperature in the rail. Thus, the non- uniform lateral extent can allow for uniform heating of the rail, to avoid certain portions becoming over- or under-heated. The frequency of the alternating current provided to the device is selected to heat various portions of the rail quicker than others. In the thinner areas of the rail, the field cancelation will be more prominent and will reduce the risks of overheating. However, in the thicker areas the field cancelation will be less and the heat will even out after the material has reached its Curie point.

Preferably, the lateral extent of the rail-facing rail-heating portion may be greater at a rail-web- corresponding portion of the longitudinal extent than at a rail-underside-of-head-corresponding portion and/or rail-top-of-foot-corresponding portion of the longitudinal extent, the rail-web- corresponding portion in use positioned at or adjacent to a web of the railway rail, the rail- underside-of-head-corresponding portion in use positioned at or adjacent to an underside of a head of the railway rail, and the rail-top-of-foot-corresponding portion in use positioned at or adjacent to a top of a foot of the railway rail. The web of the rail may require less heating than an underside of the head of the rail and the top of the foot of the railway rail to reach a welding temperature. As such, positioning the wider lateral portions here allows for balancing of the induction effect to uniformly heat the rail to a uniform temperature over time. Frequency selection also adds to this effect.

Advantageously, the lateral extent of the rail-facing rail-heating portion may be greater at a rail- side-of-head-corresponding portion of the longitudinal extent than at a rail-underside-of-head- corresponding portion and/or a rail-top-of-head-corresponding portion of the longitudinal extent, the rail-side-of-head-corresponding portion in use positioned at or adjacent to a side of a head of the railway rail, the rail-underside-of-head-corresponding portion in use positioned at or adjacent to an underside of the head of the railway rail, and the rail-top-of-head-corresponding portion in use positioned at or adjacent to a top of the head of the railway rail. The side of the head of the rail may require less heating than an underside of the head of the rail and the top of the head of the railway rail to reach a welding temperature. As such, positioning the wider lateral portions here allows for more uniform heating of the rail.

Beneficially, the tubular body comprises a tube and may further comprise lateral wings projecting from the tube to provide the non-uniform lateral extent. This provides for easier manufacturing, compared to, for example, having a tube with a non-uniform lateral extent. Additionally, it permits for the tube to have a uniform internal section which may be advantageous for fluid flow purposes. Non-uniform lateral sections may be more readily achieved with additive manufacture.

Optionally, the rail-facing rail-heating portion and/or the cooling-fluid return portion may comprise two tubes. The tubes can be spaced apart to adjust inductive heating characteristics. Additionally, two tubes for the cooling-fluid return portion allows for these to be spaced apart to maximise a separation between the rail-facing rail-heating portion and the cooling-fluid return portion. Although two tubes are preferred, it will be appreciated that more than two tubes may be considered, or only a single tube. Advantageously, a displacement of the two tubes from each other may be greater on the coolingfluid return portion than on the rail-facing rail-heating portion. This permits a greater separation between the rail-facing rail-heating portion and the cooling-fluid return portion.

Additionally, the tubular body may further comprise a joining element between the two tubes on the cooling-fluid return portion. Whilst spacing of the tubes may be advantageous, this can also result in cooling being reduced between the two tubes. Thus, providing a joining element between the two tubes, which is cooled by virtue of conduction due to its connection with the two tubes, can provide cooling between the two tubes. This may be particularly convenient for cooling induction intensifier components, if these are used, by positioning the joining element so that it engages with the induction intensifier.

In a preferable embodiment, the two tubes may be in contact on the rail-facing rail-heating portion. This can cause a more uniform temperature distribution in the rail.

Preferably, the rail-facing rail-heating portion may have a flat or substantially flat surface for presentation to the in use rail. This can cause a more uniform temperature distribution in the rail.

Advantageously, the device may further comprise at least one induction intensifier positioned at or adjacent to the copper tubular body. Induction intensifiers can direct the magnetic field towards the rail, for example in a direction of the rail-facing rail-heating portion rather than in the direction of the cooling-fluid return portion. The induction intensifier may otherwise be referred to as a flux concentrator or flux controller.

Beneficially, the or each induction intensifier may comprise ferrosilicon.

Additionally, the or each induction intensifier may be disposed between the rail-facing rail-heating portion and the cooling-fluid return portion. Such an arrangement permits for efficient cooling of the induction intensifier, whilst also providing efficient direction of the magnetic field.

In a preferable embodiment, the device may further comprise at least one heat shield arranged to shield the or each induction intensifier from radiant heat from the railway rail. The induction intensifier may have a relatively low operational temperature, becoming less effective at temperatures above around 180 °C and becoming permanently damaged at temperatures above around 350 °C. Heat shields can reduce the requirement for fluid cooling.

Beneficially, the or each heat shield may comprise ceramic. Typically, ceramics have a high melting temperature and are poor conductors of heat. Ceramic heat shields therefore provide an effective barrier to radiant heat from the rail. Typically, ceramics are electrically insulative and do not obstruct or otherwise interfere with magnetic fields. As such, the induction heating of the rail is not affected by the presence of the ceramic heat shields.

Advantageously, the copper tubular body may comprise at least one projecting fin which extends at or adjacent to the induction intensifier and from the rail-facing rail-heating portion towards the cooling-fluid return portion, or vice versa, without being in electrical contact therewith. Such a projecting fin is cooled by the fluid cooling of the tubular body and thus, in turn, cools the induction intensifier. This allows for cooling at more surface area of the induction intensifier. The projecting fin is preferably formed from the same material as the tubular body to provide efficient transfer of heat. Therefore, the projecting fin should not electrically connect the rail-facing rail-heating portion with the cooling-fluid return portion. To achieve this, an air gap may be left, or insulative material may be used between the projecting fin and one of the rail-facing rail-heating portion and the cooling-fluid return portion.

Additionally, the induction intensifier may be attached to the copper tubular body via an adhesive which is operable up to at least 150 °C and more preferably up to at least 200 °C. It may be that the induction intensifier cannot be attached to the tubular body via welding, brazing or soldering. Additionally, it may be inconvenient to form holes in the induction intensifier to permit the use of fasteners. As such an adhesive may be preferred. The adhesive may be electrically insulating, although this may not be required if the induction intensifier is not electrically conductive. The adhesive should preferably be a good conductor of heat.

Optionally, the induction intensifier may be positioned at or adjacent to the rail-facing rail-heating portion at at least one of the following portions of the longitudinal extent: a rail-top-of-head- corresponding portion which in use is positioned at or adjacent to a top of a head of the railway rail; a rail-underside-of-head-corresponding portion which in use is positioned at or adjacent to an underside of the head of the railway rail; a rail-top-of-foot-corresponding portion which in use is positioned at or adjacent to a top of a foot of the railway rail; a rail-side-of-foot-corresponding portion which in use is positioned at or adjacent to a side of the foot of the railway rail; and a rail- bottom-of-foot-corresponding portion which in use is positioned at or adjacent to a bottom of the foot of the railway rail. Such positions of the rail may require additional heating to reach a desired temperature uniformly with the remainder of the rail. As such induction intensifiers position at such portions may provide more uniform heating of the rail.

Preferably, there are a plurality of discontinuous induction intensifiers. This may allow for induction intensifiers to be used to direct the magnetic field towards specific locations of the or each rail.

In a preferable embodiment, the device may further comprise at least one insulating element comprising an electrically insulating material connected to the rail-facing rail-heating portion for electrically isolating the railway-rail induction welding device from another railway-rail induction welding device. Therefore, the two devices can be brought into engagement, which may provide for more efficient coverage of the rail, without creating electrical contact therebetween.

Advantageously, the cooling-fluid return portion may have an instep. The instep, recess, or taper may allow of the accommodation of adjacent welding system architecture or components. For example, a bar or rod which extends longitudinally alongside the rail and is required for the welding process may be accommodated in the instep. This can provide for a narrower overall apparatus, which is particularly important if welding railway rails in situations where rail to rail gaps are small, for example where rails are converging or diverging.

Beneficially, the longitudinal extent of the rail-facing rail-heating portion may be profiled to have a non-uniform separation with the rail. Thus, when a rail is received at or adjacent to the rail-facing rail-heating portion, there is a non-uniform separation therebetween. As such, there is greater inductive heating where the rail is closer to the rail-facing rail-heating portion. The irregular geometry of the rail results in certain portions of the rail requiring more heating to reach a predetermined welding temperature than other portions.

According to a second aspect of the present invention, there is provided a railway-rail induction welding apparatus comprising: two separate railway-rail induction welding devices as claimed in any one of the preceding claims, the rail-facing rail-heating portion of each device corresponding to opposing sides of a railway rail; a power source for connecting to the copper tubular body of each device; and a fluid source for connecting to the cooling-fluid inlet and cooling-fluid outlet of each device.

According to a third aspect of the present invention, there is provided a method of welding at least one railway rail, the method comprising the steps of: a) providing a railway-rail induction welding apparatus according to a first aspect of the invention; b) positioning the two welding devices at opposing sides of at least one railway rail; c) inductively heating the railway rail with the welding apparatus.

Preferably, wherein during step b), the two welding devices may be lowered at each side of the railway rail so that a surface of each welding device which during inductive heating faces the other welding device, instead faces a longitudinal direction of the rail, and then rotating the welding devices so that said surfaces face each other. The lateral extent of each device, in other words the extent in a longitudinal direction of the rail in a welding orientation, is narrower than an extent of the rail in a side-to-side direction of the rail in the welding orientation. Therefore, by rotating the device, the narrower extent of the device can be used to by-pass close system architecture.

Whilst rotating is described, it will be appreciated that other forms of movement may be considered, such as pivoting, translating or a scissor-like action.

Advantageously, each welding device may have an instep. The system architecture can be received in the instep.

The invention will now be more particularly described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 shows a front view of two railway-rail induction welding devices, arranged together as in use although without a railway rail therebetween, each an embodiment in accordance with the invention;

Figure 2 shows a side view of an upper portion of the two railway-rail induction welding devices of Figure 1 , positioned side-by-side;

Figure 3 shows a first view of an inside face of one of the railway-rail induction welding devices of Figure 1 ;

Figure 4 shows a second view of the inside face of one of the railway-rail induction welding devices of Figure 1 ;

Figure 5 shows a third view of the inside face of one of the railway-rail induction welding devices of Figure 1 ;

Figure 6 shows a bottom view of the railway-rail induction welding devices of Figure 1 positioned side-by-side;

Figure 7 shows a first view of an outside face of the railway-rail induction welding devices of Figure 1 positioned side-by-side; and

Figure 8 shows a second view of the outside face of the railway-rail induction welding devices of Figure 1 positioned side-by-side.

Referring firstly to Figure 1 , there are shown two railway-rail induction welding devices 10, inductors or coils. The two railway-rail induction welding devices 10 are arranged so as to face each other, defining a space 12 therebetween in which in use at least one railway rail to be inductively heated would be received.

Each welding device 10 comprises a copper tubular body 14 having a cooling-fluid inlet and a cooling-fluid outlet. The cooling-fluid inlet and outlet are not shown in the figures; however, the cooling-fluid inlet is in a rail-proximal side 16a of an electrical connection flange 16, and the cooling-fluid outlet is in a rail-distal side 16b of the electrical connection flange 16.

The copper tubular body 14 defines at least one continuous conduit from the cooling-fluid inlet to the cooling fluid outlet. The conduit is for water to flow along and thereby cool the tubular body 14. The water is preferably at a temperature of up to 32 °C, in particular 12 °C or 13 °C, and may flow at a rate of between 10 and 30 litres per minute, in particular 18 litres per minute. However, it will be appreciated that other temperatures and other flow rates may be considered. The copper tubular body 14 also defines a continuous electrical path, circuit or connection. This electrical path is preferably at least along the walls of the continuous conduit. The electrical path is between the rail-proximal side 16a of the electrical connection flange 16 and the rail-distal side 16b of the electrical connection flange 16, since this is where terminals of a power source are connected.

The copper tubular body 14 preferably comprises, consists essentially, or consists of copper. Such copper is preferably oxygen-free copper. However, it will be appreciated that alloys or other grades of copper may be considered, and/or that parts of the copper tubular body 14 may be brazed or soldered together using a filler which comprises other elements. Additionally, if the copper tubular body 14 is formed via additive manufacture, then other materials may be included with the copper to facilitate such manufacture, although additive manufacture material can be 99.9% copper. Although copper is described for the tubular body 14, it will be appreciated that other highly conductive materials could be considered, such as silver or gold.

The copper tubular body 14 has a rail-facing rail-heating portion 18 and a cooling-fluid return portion 20.

The copper tubular body 14 has an upstream connection portion 22 between the cooling fluid inlet and the rail-facing rail-heating portion 18, and a downstream connection portion 24 between the cooling-fluid return portion 20 and the cooling fluid outlet. As such, cooling fluid is configured to flow from the cooling fluid inlet, through the upstream connection portion 22, the rail-facing railheating portion 18, the cooling-fluid return portion 20, downstream connection portion 24, and out the cooling fluid outlet. Similarly, electrical current flows from the rail-proximal side 16a of the electrical connection flange 16 to the rail-distal side 16b of the electrical connection flange 16 via the upstream connection portion 22, the rail-facing rail-heating portion 18, the cooling-fluid return portion 20, and the downstream connection portion 24, and vice versa.

The upstream connection portion 22 and downstream connection portion 24 are mounted to a spacing or support structure 26. The spacing structure 26 comprises an upstream backing plate 26a, for the upstream connection portion 22, and a downstream backing plate 26b for the downstream connection portion 24. The backing plates 26a, 26b space the electrical connection flange 16 from the rail-facing rail-heating portion 18 in a vertical direction.

The plates 26a, 26b are here made from copper and are electrically isolated from each other. The plates 26a, 26b are electrically isolated via an insulating plate 28 or element, which may be formed from ceramic, for example. The backing plates 26a, 26b and insulating plate 28 are here engaging and joined together for structural rigidity, the joining being achieved by fasteners 28a, shown in Figure 2, which are made from an electrically insulating material, such as ceramic. It will be appreciated that the plates may be considered to be part of the tubular body itself, since they are formed of copper. A non-copper spacing structure may also be considered, particularly if formed from an electrical insulator.

The spacing structure 26 also comprises a base 26c which extends or projects from the upstream backing plate 26a towards the centre of the or each rail. The base 26c projects at substantially 90 degrees from the backing plate. The upstream connection portion 22 extends down the upstream backing plate 26a and then across the base 26c so that it is adjacent to the centre of the rail. The downstream connection portion 24 extends up the downstream backing plate 26b. There are preferably two L-shaped or substantially L-shaped braces 30 between the base 26c and the upstream backing plate 26a to provide structural integrity and support.

The rail-facing rail-heating portion 18 has a longitudinal extent which is non-linear and profiled to correspond or substantially correspond to one side of the railway rail to be treated or heated. The longitudinal extent will be understood to mean the direction which extends along or parallel to the flow path of fluid or electricity through the tubular body 14. This is effectively in a top-to-bottom direction along the side of the or each railway rail.

A railway rail typically has a head, a web and a foot. As such, the rail-facing rail-heating portion 18 is profiled to correspond to this shape. The longitudinal extent of the rail-facing rail-heating portion 18 therefore has a rail-top-of-head-corresponding portion 32 which in use is positioned at or adjacent to a top of a head of the railway rail, rail-side-of-head-corresponding portion 34 which in use in use is positioned at or adjacent to a side of the head of the railway rail, a rail-underside- of-head-corresponding portion 36 which in use is positioned at or adjacent to an underside of the head of the railway rail, a rail-web-corresponding portion 38 which in use is positioned at or adjacent to a web of the railway rail, a rail-top-of-foot-corresponding portion 40 which in use is positioned at or adjacent to a top of a foot of the railway rail, a rail-side-of-foot-corresponding portion 42 which in use is positioned at or adjacent to a side of the foot of the railway rail, and a rail-bottom-of-foot-corresponding portion 44 which in use is positioned at or adjacent to a bottom of the foot of the railway rail.

The rail-facing rail-heating portion 18 has a lateral extent which is non-uniform along the said longitudinal extent of the rail-facing rail-heating portion 18. In other words, the copper tubular body 14 at the rail-facing rail-heating portion 18 has a variable width or breadth. The lateral direction is preferably defined as the direction in a longitudinal direction or axial direction of the rail. However, the lateral direction may additionally or alternatively be defined as the side-to-side direction of the rail, and in this case, a varying thickness of the tubular body may define the non-uniform lateral extent.

Referring to Figure 3, preferably, the lateral extent of the rail-facing rail-heating portion 18 is greater at the rail-web-corresponding portion 38 than at the rail-underside-of-head-corresponding portion 36 and/or the rail-top-of-foot-corresponding portion 40. Referring to Figure 4, additionally or alternatively, the lateral extent of the rail-facing rail-heating portion 18 is greater at the rail-side- of-head-corresponding portion 34 than at the rail-underside-of-head-corresponding portion 36 and/or the rail-top-of-head-corresponding portion 32. Most preferably, at the rail-web- corresponding portion 38 and/or the rail-side-of-head-corresponding portion 34, the lateral extent is the greatest compared to any other portion of the rail-facing rail-heating portion 18.

In use, the additional lateral extent at these portions 38, 34 of the copper tubular body 14 reduces the density or intensity of the induction or inductive field at the corresponding locations in the rail. This thus results in a more uniform heating of the rail, since otherwise these locations of the rail would overheat and melt or soften in advance of the other locations. Such locations may be more susceptible to reaching a high temperature when inductively heated, due to their geometry. It will be appreciated that for other rail geometries or copper tubular bodies with other shaped profiles, other portions of the rail-facing rail-heating portion may have differing lateral extents.

The tubular body 14 comprises at least one tube for defining the conduit for the cooling fluid. Preferably there are two tubes 46a, 46b along the rail-facing rail-heating portion 18 and the cooling-fluid return portion 20. The tubes 46a, 46b are separate and define separate conduits along the rail-facing rail-heating portion 18 and the cooling-fluid return portion 20, although it will be appreciated that that the tubes may merge at some locations along the rail-facing rail-heating portion 18 and the cooling-fluid return portion 20 if desirable. At the rail-facing rail-heating portion 18, the tubes 46a, 46b are preferably close together, and more preferably are engaging.

Along the rail-facing rail-heating portion 18 the or each tube 46a, 46b preferably has a polygonal cross-section and/or may have at least one flat side so as to provide a flat surface adjacent to the rail. For example, here square or rectangular tubes are used. Together, the tubes have a 20 mm width. Here this is achieved by using two 10 mm square tubes. Other dimensions may be considered.

Along the rail-facing rail-heating portion 18, the diameter of the or each tube 46a, 46b is preferably constant. Therefore, to achieve the greater lateral extent, side plates 48 or wings are brazed, soldered, or otherwise formed onto at least one side, and preferably both sides, of the tube or both tubes 46a, 46b. This can be seen in Figures 3, 4 and 5. However, it will be appreciated that other arrangements may be considered for obtaining the greater lateral extent. For example, using tubes with varying or non-uniform diameter.

Along the cooling-fluid return portion 20, the tubes 46a, 46b may similarly have a 10 mm diameter and/or may be square or rectangular shaped, or rounded. However, other dimensions and shapes may be considered. The tubes 46a, 46b along the cooling-fluid return portion 20 are spaced apart or separated to maximise the separation between the cooling-fluid return portion 20 and the rail- facing rail-heating portion 18. This can be seen in Figures 6, 7 and 8. The spacing is such that each tube of the cooling-fluid return portion 20 may be considered to be offset from the tubes of the rail-facing rail-heating portion 18 by 45 degrees.

Along the upstream and downstream connection portions 22, 24 the copper tubular body 14 preferably comprises a single tube, for example a 20 mm by 10 mm rectangular tube. As such, T-junctions 50 or splitters, as shown in Figure 2, may be used in the tubing between the upstream connection portion 22 and the rail-facing rail-heating portion 18, and the downstream connection portion 24 and the cooling-fluid return portion 20. The tubular body 12 may be described as a manifold. The tubular body 12 is the electrical node point. Thus, the tubular body may be described as a manifold node.

Although tubes of simple shapes and dimension are described, it will be appreciated that more complex tube or conduit configurations or shapes may be considered. For example, hollow tubes with a geodetic structure or more than two small tubes formed into a singular tubular format. Additive manufacture in particular may permit for such complex shapes.

The rail-facing rail-heating portion 18 substantially corresponds to one side of a railway rail and is configured to be spaced apart therefrom, preferably and generally between 3 mm and 10 mm and in particular 6 mm. However, this separation is preferably non-uniform and as such there is less separation between the rail-facing rail-heating portion 18 and the railway rail at certain positions where a more intense induction heating action is required. For example, at or adjacent to positions where the railway rail is thickest. As can be seen in Figures 3, 4 and 5, to achieve a closer separation, copper plates 52 are joined, for example by brazing, soldering or otherwise joining, to the surface of the tubes 46a, 46b. This is at or adjacent to the rail-top-of-head- corresponding portion 32, preferably towards to the centre of the rail, the rail-underside-of-head- corresponding portion 36, and the rail-bottom-of-foot-corresponding portion 44, preferably towards to the centre of the rail. Whilst brazing or soldering a copper plate to the tubes is described, it will be appreciated that instead the tubes may be formed, such as being bent, into the desired shape to achieve the closer separation.

Referring again to Figure 1 , to increase induction in the rail at specific locations, induction intensifiers 54 may be positioned at or adjacent to specific portions of the rail-facing rail-heating portion 18. The induction intensifiers 54 may direct the magnetic field towards the or each rail, for example in a direction of the rail-facing rail-heating portion 18 rather than in a direction of the cooling-fluid return portion 20, and in particular towards specific locations of the or each rail. The induction intensifiers 54 are preferably disposed between the rail-facing rail-heating portion 18 and the cooling-fluid return portion 20 of the tubular body 14. The induction intensifiers 54 are secured to the rail-facing rail-heating portion 18 and/or the cooling-fluid return portion 20 with adhesive. The adhesive is preferably operable up to at least 150 °C and more preferably up to at least 200 °C. The adhesive should be a good conductor of heat. Suitable adhesive is a 2 pack epoxy resin, for example that supplied by Araldite (RTM).

The induction intensifiers 54 are here positioned at or adjacent to portions of the longitudinal extent to increase the induction in the or each rail at the corresponding positions in the or each rail. The portions with an induction intensifier 54 positioned at or adjacent thereto are preferably the rail-top-of-head-corresponding portion 32, the rail-underside-of-head-corresponding portion 36, the rail-top-of-foot-corresponding portion 40, the rail-side-of-foot-corresponding portion 42, and/or the rail-bottom-of-foot-corresponding portion 44. However, it will be appreciated that one or more such portions may be omitted or other portions may be considered.

The induction intensifiers 54 comprise, consist essentially of, or consist of ferrosilicon. Suitable material can be obtained from POCO Magnetic Co., Ltd, China. However, other induction intensifier 54 materials may be considered. For example, magnetic laminations, iron oxides, ferrites including soft ferrites or other ferrimagnetic materials including nickel-based/comprising or cobalt-based/comprising ferrimagnetic materials.

The induction intensifiers 54 are preferably discontinuous. In other words, they are separate pieces. However, it will be appreciated that a continuous induction intensifier may be considered.

The device 10 preferably further comprises at least one heat shield 55 for protecting components of the device 10 from melting or softening due to radiant heat from the rail. The or each heat shield 55 is for the induction intensifiers 54 and is positioned at or adjacent to the rail-facing railheating portion 18. It will be appreciated that heat shielding may also be used for the copper tubular body.

The or each heat shield 55 is formed from an insulating and non-conductive material, particularly a ceramic material although other materials may be considered. Suitable ceramic material should be able to withstand 1200 °C and can be obtained from Esspee, for example Zircar.

There is preferably a plurality of spaced apart heat shields 55, although a single continuous heat shield 55 may be used. The or each heat shield 55 preferably is positioned either side of the copper tubular body 14, in a lateral direction, and at or adjacent to the induction intensifiers 54, so as to not overlap a rail facing surface of the copper tubular body 14 whilst still providing a heat shielding effect to the induction intensifiers 54. The heat shields 55 also preferably extend or project laterally beyond the induction intensifiers 54, in an axial direction of the rail, so as to provide additional lateral heat protection.

Although a separate heat shield 55 is described, it will be appreciated that other arrangements may be considered, such as a heat resistant coating, for example a ceramic coating, applied to the induction intensifiers 54. Additionally, if fluid cooling can be sufficiently applied to the induction intensifiers 54, and/or if induction intensifiers 54 with high enough melting temperatures can be used, then heat shields 55 may be omitted entirely.

To assist with cooling the induction intensifiers 54, cooling fins or plates 56 extend or project from the tubular body 14 and engage with the induction intensifiers 54. This may be in an axial or longitudinal direction of the or each rail, for example extending from one of the tubes 46a of the cooling-fluid return portion 20 towards the other tube 46b of the cooling-fluid return portion 20. The cooling fin or plate 56 may extend between the two tubes of the of the cooling-fluid return portion 20. This can be seen in Figures 6, 7 and 8.

Additionally, or alternatively, cooling fins or plates may extend or project from the tubular body 14, engaging with the induction intensifiers 54, in a direction from the rail-facing rail-heating portion 18 towards the cooling-fluid return portion 20, and/or vice versa. In such a case, the cooling fins or plates preferably should not create electrical contact between the rail-facing railheating portion 18 and cooling-fluid return portion 20. Therefore, in the instance that the cooling fins are electrical conductors, there is a gap or insulation between the cooling fin or plate and one of the rail-facing rail-heating portion 18 and cooling-fluid return portion 20.

The cooling fins or plates are preferably formed from the same material as the copper tubular body 14, and may be integrally formed therewith, or joined thereto, for example via brazing or soldering.

The base 26c of the spacing structure 26 extends at or adjacent to the induction intensifier 54 at the rail-top-of-head-corresponding portion 32. As such, it is advantageous to have the backing plates 26a, 26b sited at an outside edge of the device 10, to permit cooling fluid at a lower temperature due to its proximity to the inlet to extend further across the induction intensifier 54.

The cooling-fluid return portion 20 preferably includes an instep 58, depression, or taper in the in use side-to-side direction of the or each rail. This permits for the accommodation of architecture such as axially extending bars and therefore allows use of the device 10 in a narrower space. However, if use in a narrower space were not required, then the instep may not be necessary, and the cooling-fluid return portion 20 could extend straight, smoothly or uniformly to the coolingfluid outlet.

As shown in Figure 1 , in use two devices 10 are brought together. Electrical contact should be avoided between the two devices 10, and therefore electrically insulating material is positioned between the two devices 10, above and/or below the rail. Such insulating material is provided on upper and lower closure plates 60, although this is not shown in the figures. The insulating material is preferably mica, although other electrically insulating materials having a high melting temperature may be considered. The insulating material is preferably joined to the closure plates 60 with a high temperature adhesive or high temperature tape, such as ceramic tape. The tape is preferably non-conductive. However, it will be appreciated that it may be possible to maintain an air gap between the two devices 10, in which case it may be possible to avoid the use of electrically insulating material disposed between the devices 10. Alternatively, the same adhesive as used for the induction intensifiers can be used.

In use, two devices 10 are provided and may be movably mounted to a support structure. For each device 10, one terminal of an alternating current power supply, which may be a handheld transformer, is connected to the rail-proximal side 16a of the electrical connection flange 16 and one terminal is connected to the rail-distal side 16b of the electrical connection flange 16. A cooling fluid source, such as water at a temperature of up to 32 °C flowing at a rate of 18 litres per minute, is connected to the cooling-fluid inlet. A drain is connected to the cooling-fluid outlet. Such an arrangement provides a railway-rail induction welding apparatus.

The or each railway rail to be heated and welded is raised from the ground and each device 10 is preferably vertically lowered either side of the or each railway rail. For example, for defect removal, the devices 10 are lowered either side of the rail at a location where a defect is located.

The devices 10 are preferably lowered in a transverse or perpendicular orientation to an orientation of the devices 10 when heating the rail. In other words, the devices 10 are lowered so that the outward surface 62 of the rail-facing rail-heating portion 18 of each rail faces the axial direction of the rail. This permits the narrower lateral extent of the devices 10 to pass by the axially extending supporting bars. The devices 10 are then rotated so that the axially extending supporting bars are received or accommodated in the instep 58. The outward surface 62 of the rail-facing rail-heating portion 18 of one device 10 thus faces the outward surface 62 of the railfacing rail-heating portion 18 of the other device 10.

The devices 10 are then brought into contact with each other at the closure plates 60, although electrical isolation is maintained by the mica plate. Fluid and alternating electric current are provided through the copper tubular body 14 of each device 10. The alternating electric current in each device 10 creates an alternating magnetic field around each device 10 which inductively heats the railway rail located between the two magnetic fields. The induction intensifiers 54 direct the alternating magnetic field towards the railway rail. The non-uniform lateral extent of the railfacing rail-heating portion 18, and the non-uniform separation between the rail and the tubular body 14, provide uniform heating in the railway rail. The copper tubular body 14 is protected from melting despite resistance heating and/or radiant heat from the railway rail due to the cooling fluid. The induction intensifiers 54 are protected from melting by the heat shields 55.

The railway rail is softened and/or melted by the induction heating. The two portions of the railway rail surrounding the defect can then be urged towards each other, effectively removing the defect. A shear die can be used to remove weld material which has erupted from between the rails. The electric current can then be terminated if not already done so and the devices 10 removed from the rail.

If the welding apparatus is being used for joining two railway rails, the two railway rails may be aligned and the welding apparatus positioned around an end portion of one or both of the railway rails in a similar way as previously described. The or each said end is inductively heated so as to soften and/or melt and the two railway rails urged together, welding the rails. The devices 10 can then be removed from the joined rail. A shear die can be used to remove weld material which has erupted from between the rails.

Although not shown in the figures, it will be appreciated that at least one pyrometer, or other temperature gauge, may be used to measure the surface temperature of the rail, or that of the induction intensifier 54, whilst the rail is being inductively heated. This allows monitoring of the temperature of the rail and/or induction intensifier 54 to determine whether the power supplied to the tubular body 14 should be adjusted to avoid over or under heating.

To allow access for the or each pyrometer, at least one aperture should be present in the railfacing rail-heating portion 18. The or each aperture preferably has a diameter of between 2 mm and 20 mm, and more preferably may have a diameter of 5 mm or 15 mm. The or each aperture preferably has a circular cross-section. There may be differently sized apertures in the rail-facing rail-heating portion 18 to accommodate different pyrometers.

Any of the following portions of the longitudinal extent of the rail-facing rail-heating portion 18 may have an aperture therein: the rail-top-of-head-corresponding portion 32, the rail-side-of-head- corresponding portion 34, and/or the rail-web-corresponding portion 38. In the rail-top-of-head- corresponding portion 32, the aperture may be defined by the two devices 10 together, and therefore each may define a half-aperture. In other words, each may have a semi-circular cut-out or groove at the edge of the rail-top-of-head-corresponding portion 32.

Additionally or alternatively, at least one cooling plate at the cooling-fluid return portion 20 may have an aperture therein to allow pyrometer access to the induction intensifier 54 which may otherwise be obscured by the cooling plate. This may, for example, be to measure the temperature of the induction intensifier 54 at the rail-side-of-foot-corresponding portion 42.

It is therefore possible to provide railway-rail induction welding device for inductively welding at least one rail. The induction welding device has a rail-facing rail-heating portion which has varying width to provide uniform heating of the irregular geometry of the rail.

The words ‘comprises/comprising’ and the words ‘having/including’ when used herein with reference to the present invention are used to specify the presence of stated features, integers, steps or components, but do not preclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination.

The embodiments described above are provided by way of examples only, and various other modifications will be apparent to persons skilled in the field without departing from the scope of the invention as defined herein.