REPAIR METHOD AND ALLOY

Field of the Invention

The present invention relates to a method for repairing an imperfection in a metallic article and to a bismuth based alloy which may suitably be used in the method.

Background to the Invention Equipment used in engineering processes which have passageways for passage of heat exchange fluids can form imperfections such as cracks and micro-cracks as the result of the high temperatures and forces placed upon them. Typical of such equipment are casting moulds, casting dies, injection moulding dies and heat exchangers. The formation of these cracks and micro-cracks result in a variety of problems for the engineering process . In particular, if cracks develop in the walls of cooling or heating fluid conduits, the cooling fluid may leak. This may result in the cooling fluid interacting with the function of the apparatus, eg. fluid entering a die cavity, loss of thermal control and/or a reduction in the thermal efficiency of a casting die.

Commonly, these imperfections are repaired by inserting a neat fitting tube into the conduit and connecting it to the cooling or heating fluid supply. However, the thermal performance of the equipment can be greatly reduced due to the gaps which exist between the external diameter of the inserted tube and the internal diameter of the passage way. Attempts have been made to overcome the problem of these gaps by various techniques such as peening, shrink fitting or using fillers such as solder. However, none of these techniques has resulted in satisfactory thermal performance of the required passageway.

Another problem with the method of repairing cracks using a neat fitting tube is that it cannot be used

in multiple conduits in a circuit made from intersecting drilled holes as it does not provide an adequate seal to prevent leakage of fluid.

Summary of the Invention

In a first aspect, the present invention provides a method for repairing an imperfection in a metallic article, the method comprising the steps of applying a molten alloy which expands upon solidification to the imperfection, and allowing the applied alloy to solidify.

The method may further comprise the step of shaping the solidified alloy. When shaped, the solidified alloy will typically be shaped by a machining operation such as drilling. Typically, the imperfection will be a crack or the like in a heat exchange conduit of the article. In applying the molten alloy to the imperfection, it is to be noted that the molten alloy need not fill the imperfection. Rather, the alloy may form a skin which spans the imperfection. It is to be noted that the imperfection may be a hairline crack only microns in width.

Typically, the article will be a casting mould or casting die and the method will be used to restore, or substantially improve, the heat transfer performance of the article.

The method may further comprise preheating the article prior to applying the molten alloy.

The molten alloy may be applied under gravitational force, under an applied force or under vacuum. Typically, the step of applying the molten alloy will comprise casting the alloy into a conduit in the article. The alloy may be cast to fill substantially the entire conduit or only a portion of the conduit. If only a portion of the conduit is to be filled with molten alloy, then preferably, the method further comprises the step of inserting a removable plug into the conduit prior

to the step of applying molten alloy to the imperfection.

The method may further comprise the step of inserting a tube into a conduit prior to applying the imperfection with the molten alloy between the outside wall of the tube and the inside wall of the conduit .

The tube may comprise a number of projections on its outer wall to enable the tube to be centred within a conduit prior to applying the imperfection with the molten alloy. In an embodiment, the projections are laterally spaced apart along the length of each tube so as to not provide any significant obstruction to the casting of the alloy.

The projections may be deposits of weld or a brazing metal or solder, or small buttons attached by welding, brazing, soldering or by any other suitable fixing method.

In another embodiment, the projections may be formed by machining away a layer of metal of the tube to leave free standing studs, spigots or stand offs.

In an embodiment where the conduit is a blind hole, the tube will typically comprise a sealed end. The sealed end may also comprise a projection projecting therefrom, to space the sealed end away from the blind end of the conduit.

In an embodiment, the step of applying the molten alloy involves casting the alloy into two intersecting conduits in the article. The conduits may or may not intersect at right angles . In this embodiment, the method may further comprise the steps of inserting a first tube into the first conduit and a second tube into the second conduit. The first and second tubes may or may not be joined together . The method may comprise the step of joining the first and second tubes together.

In one embodiment, the first tube is formed with

a taper threaded end and the second tube is drilled and tapped to produce an aperture of appropriate size and orientation to receive the taper threaded end of the first tube. In this embodiment, the steps of inserting the first and second tubes into the first and second conduits also involves simultaneously joining the first tube to the second tube by receiving the taper threaded end of the first tube in the aperture of the second tube. Molten alloy is subsequently cast between the outside wall of the tubes and the inside walls of the conduits.

In another embodiment, the method further comprises inserting a shaped plug into one end of the first tube, the distal end of the shaped plug being shaped to fit with the contour of the outer surface of the second tube.

In this embodiment, portions of the surfaces of the plug and the first and second tubes in the area at and around which the join is to be formed between the tubes may be coated with a suitable soldering flux. Preferably, the soldering flux has an operating temperature that encompasses the melting temperature of the alloy. Preferably, the soldering flux is suitable for use with steel, copper and copper alloys. Typically, the application of soldering flux facilitates a metallurgical bond forming between the surface of the tubes and the alloy surrounding the joint, thereby increasing the strength of the joint. An example soldering flux is a zinc-chloride flux.

The method of this embodiment further comprises the step of inserting the tubes into the conduits, with the distal end of the shaped plug snugly abutting the second tube, prior to applying the imperfection with the molten alloy between the outside wall of the tubes and the inside wall of the conduits. Once the first and second tubes have been joined together, the method further comprises the step of drilling through the shaped plug and the outer wall of the

second tube to fluidly connect the first and second tubes together .

The alloy expands upon solidification and may be, for example, commercially available bismuth-tin alloys or lead containing alloys such as solders, although lead containing alloys may be undesirable due to occupational health and safety concerns. Preferably, the alloy is an alloy according to the second aspect of the present invention discussed below. In a second aspect, the present invention provides a bismuth based alloy consisting of 0.05 - 10% by weight zinc, 0 - 1%, preferably 0%, by weight magnesium, 0 - 3%, preferably 2 - 3%, by weight silver and the balance being bismuth except for incidental impurities. A particularly preferred composition is a eutectic mixture of bismuth and zinc containing about 2.7% by weight zinc and, optionally, about 2.5% by weight silver.

Pure bismuth metal expands upon solidification. However, as it solidifies, large grains are typically formed due to the absence of grain boundary pinning and solute concentration gradients . These large grains produce a solidified metal that is brittle and low in yield strength. Typically, the solidified metal is therefore easily damaged when machined. Magnesium and silver are optional components of the alloy. When present, these elements improve the strength and ductility of the alloy. When present, magnesium and silver have a grain refining effect on the alloy. However, magnesium is preferably not included in the alloy as it may increase the susceptibility of the alloy to corrosion.

It has been surprisingly found that by alloying bismuth with small amounts of zinc, magnesium and silver, an alloy is produced which has reduced grain size when cast, thus improving the strength and machinability of the cast metal as compared with pure bismuth, whilst retaining pure bismuth's property of expanding upon solidification.

Without wishing to be bound by theory, the component elements of the alloy are believed to form binary, ternary and quaternary eutectics thus resulting in the improved hardness and tnachinability of the alloy over pure bismuth.

Alloys according to the second aspect of the present invention have a melting temperature in the range of 240° to 250 ⁰ C. The alloy melt temperature is thus well above the boiling point of water, yet well below the melting temperature of common engineering structural materials such as steel, cast iron, copper based alloys, aluminium based alloys and zinc based alloys from which casting die blocks, for example, are manufactured. This physical property of the alloys enables them to be used as a solid in contact with low temperature heat exchange fluids, such as water, and to be applied as molten alloy to surfaces of metallic articles.

Due to the large differences in the densities, melting points and oxidation resistances of bismuth, zinc and magnesium, a number of problems exist in alloying these metals. In particular, bismuth melts at a temperature ( 271.4°C) well below the melting points of zinc (419°C) , silver (692°C) and magnesium (651°C) , and both zinc and magnesium have densities such that in solid form they both float on the surface of molten bismuth.

The present inventors have found that solid zinc can be added to molten bismuth at temperatures above 300 ⁰ C, for example, 360-400 ⁰ C, if the zinc is located below the surface of the molten bismuth. Similarly, it has been found that solid magnesium can be added to molten bismuth or a molten bismuth-zinc alloy at a temperature in the high 300 ⁰ C range, for example, 370-400 ⁰ C, by holding the magnesium below the melt surface. Silver is preferably added first to molten zinc, with subsequent addition of the silver-zinc alloy to molten bismuth. This is generally to avoid heating the bismuth to excessive temperatures. In an embodiment, a zinc-silver alloy is formed according to

the steps of heating zinc to between 650 ⁰ C prior to adding molten silver to the zinc.

Alternatively, a combination of bismuth-zinc, zinc-magnesium and/or zinc-silver master alloys may be separately formed and then combined to achieve the desired alloy composition.

In a further alternative approach, the zinc, magnesium and/or silver alloying components may be added, in whole or in part, to molten bismuth in the molten state.

According to a third aspect of the present invention, there is provided a method for forming a bismuth based alloy according to the second aspect of the present invention comprising the steps of melting bismuth to a temperature of above 300 ⁰ C and adding the zinc content of the alloy to the molten bismuth below the surface of the molten bismuth.

The bismuth may be melted to a temperature of 360-400°C. The method for forming the bismuth based alloy may further comprise adding the zinc content in solid form.

The method may further comprise holding the solid zinc content below the melt surface. Where the alloy has a magnesium content, the method may further comprise the step of adding the magnesium content to the molten bismuth below the surface of the molten bismuth.

The method may further comprise adding the magnesium content in solid form.

The method may further comprise holding the solid magnesium content below the melt surface.

Where the alloy has a silver content, the method further may comprise forming a zinc-silver alloy prior to adding the silver content with the zinc content to the molten bismuth by adding the zinc-silver alloy to the molten bismuth.

When forming the bismuth based alloy, it is preferable to provide a controlled atmosphere. Where the alloy is magnesium and silver free, the controlled atmosphere may be an inert atmosphere such as nitrogen in order to reduce the formation of dross . For alloys containing magnesium, a cover gas of the kind used in magnesium foundry practice is preferably blanketed over the melt. Similarly, for alloys containing silver, a suitable cover gas is blanketed over the melt. According to a fourth aspect of the present invention, there is provided a method for improving the heat transfer between a hot body and a cold body in an article, the method comprising the step of casting a molten alloy in a gap between the bodies . The alloy is preferably the bismuth based alloy according to the second aspect of the present invention.

The article may be an injection moulding die or an electronic device such as a laptop computer, for example . The cold body may be a cooling mechanism such as heat pipes, for example.

The hot body may be a heat load or heat sink, for example .

Brief Description of the Drawings

Preferred embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

Figures IA, IB and 1C are schematic views of a first method for repairing a crack in a fluid conduit of a casting die;

Figures 2A and 2B are schematic views of a second method for repairing a crack in a fluid conduit of a casting die; Figures 3A, 3B and 3C are schematic views of further methods for repairing a crack in a fluid conduit of a casting die;

- S -

Figures 4A and 4B are schematic views of a further alternative method for repairing a crack in a fluid conduit of a casting die,-

Figures 5A, 5B and 5C are schematic views of a yet further alternative method for repairing a crack in a fluid conduit of a casting die;

Figures 6A and 6B are schematic views of a cylindrical test apparatus on which repair methods according to embodiments of the present invention were used;

Figures 7, 8 and 9 are graphs of data obtained from measuring the increase in water temperature as it flowed through the test apparatus of Figures 6A and 6B;

Figures 1OA and 1OB are schematic views of a further method for repairing a crack in a fluid conduit of a casting die;

Figures HA and HB are schematic views of another method for repairing a crack in a fluid conduit of a casting die; Figures 12A and 12B are schematic views of yet another method for repairing a crack in a fluid conduit of a casting die,- and

Figures 13A and 13B are schematic views of a further method for repairing a crack in a fluid conduit of a casting die.

Detailed Description of a Preferred Embodiment

Referring firstly to Figures IA, IB and 1C, a casting die block 10 is shown having a conduit 11 which under normal operation of the casting die block 10 carries a cooling or heating fluid. The casting die block 10 has an imperfection in the form of a crack 12, extending from the wall of the casting die block to the inner wall of the conduit 11. If not repaired, this crack 12 may lead to a variety of operational problems for the casting block 10, including leaking of the fluid from the conduit 11 into the casting die, and/or a general loss of thermal

efficiency of the casting die.

A method of repairing the crack 12 is shown sequentially in Figures IA, IB and 1C. Figure IA shows the cracked casting die block 10. In Figure IB, the conduit 11 is filled with a molten alloy 20 of nominal composition 96.8% by weight bismuth, 2.7% by weight zinc and 0.5% by weight magnesium. Upon solidification, the alloy expands to form a seal along the inner wall of the conduit 11, and hence across the crack 12. Prior to being filled with molten alloy, the conduit 11 is cleaned and casting die block 10 is preheated to assist in the flow of the molten alloy 20 into the conduit 11.

The conduit 11 may be chemically cleaned or fluxed and/or mechanically cleaned by, for example, wire brushing, to remove any scale, corrosion or other loosely adhering materials from the inner wall of the conduit 11.

Figure 1C shows the next step in the method for repairing the crack 12. This involves machining, typically by drilling, a new conduit 21 through the alloy 20 once it has solidified in the conduit 11. The action of drilling out the new conduit 21 leaves a thin wall of alloy 20 over the inner wall of the original conduit 11. Due to the expansion of the alloy 20 as it solidifies, this thin wall provides a seal across the entire wall of the conduit 11 including the crack 12. As a result, the casting die block 10 has been repaired to substantially prevent any leakage of fluid from the new conduit 21 without suffering from any of the problems associated with the prior art repair techniques, in particular the loss of thermal efficiency. This is because there is good thermal transfer from the casting die block 10 to the thin wall of alloy 20 and subsequently to any cooling or heating fluid in the new conduit 21.

Referring now to Figures 2A and 2B, an alternative method for repairing a crack 112 in a casting die block 110 is shown sequentially. Equivalent features in Figures 2A and 2B to those shown in Figures IA to 1C

have been given the same reference number, but include the prefix numeral 1.

The first step in this alternative method, is shown in Figure 2A. A tubular insert 122 formed from copper alloy is positioned inside the conduit 111. The insert 122 does not have to extend beyond the length of the conduit as shown in Figure 2A. The second step in this alternative method is depicted in Figure 2B. An alloy 120 is cast between the outer wall of the tube insert 122 and the inner wall of the conduit 111. The alloy 120 expands upon solidification, thus forming a tight seal across the outer wall of the tube insert 122 and across the inner wall of the conduit 111 and hence over the crack 112. Thus, the alloy 120 provides for good thermal transfer from the casting die block 110 through to the tube insert 122 and hence to any heating or cooling fluid flowing through the tube insert 122 during operation of the casting die 10.

Referring now to Figures 3A, 3B and 3C, a further alternative method for repairing a crack 212 in a casting die block 210 is shown. Similar features in Figures 3A to 3C to those in Figures IA to 1C have been given the same reference number, but are prefixed with the numeral 2. The method for repairing the crack 212 shown in Figures 3A to 3C, is of particular application where it is desirable only to repair a portion of a conduit 211.

The first step of this alternative method, as shown in Figure 3A, involves inserting a removable plug 223 into the conduit 211. The plug 223 does not cover the crack 212. The crack 212 can then be repaired by casting an alloy 220 into the conduit 211 up to the removable insert 223 and across the crack 212, and then subsequently drilling a new conduit 221 through the alloy 220, as shown in Figure 3B. Alternatively, as shown in Figure 3C, the crack 212 could be repaired by placing a tubular insert 222 in the conduit 211 and casting an alloy 220 between the outer wall of the tube insert 222 and the inner wall

of the conduit 211. The alloy 220 expands upon solidification, thus avoiding a substantial loss in thermal efficiency of the casting die 210, whilst satisfactorily repairing the crack 212 so as to substantially prevent leaking of fluid from the conduit 211. Once the alloy 220 has solidified, in both of the aforementioned cases, the plug 223 can be removed from the conduit 211.

Referring now to Figures 4A and 4B, a further alternative method for repairing a crack 412 in a casting die block 410 is shown. Similar features in Figures 4A and 4B to those in Figures IA to 1C have been given the same reference number, but are prefixed with the numeral 4. The method for repairing the crack 412 shown in

Figures 4A and 4B, is of particular application where it is desirable to repair a conduit 411 that is a blind hole in die block 410.

The first step of this alternative method, as shown in Figure 4A, involves inserting a tube with a sealed end into the blind conduit 411.

A tubular insert 422 formed from a suitable metal tube with one end sealed by a permanent plug, is positioned inside the conduit 411. The insert 422 does not have to extend beyond the length of the conduit as shown in Figure 4B. The second step in this alternative method is depicted in Figure 4B. An alloy 420 is cast between the outer wall and the bottom of the tube insert 422 and the inner wall of the conduit 411. The alloy 420 expands upon solidification, thus forming a tight seal across the outer wall of the permanently plugged tube insert 422 and across the inner wall of the conduit 411 and hence over the crack 412. Thus, the alloy 420 provides for good thermal transfer from the casting die block 410 through to the tube insert 422 and hence to any heating or cooling fluid flowing through the tube insert 422 during operation of the casting die 410.

The alloy 420 expands upon solidification, thus avoiding a substantial loss in thermal efficiency of the casting die 410, whilst satisfactorily repairing the crack 412 so as to substantially prevent leaking of fluid from the conduit 411.

Referring to Figures 5A, 5B and 5c, a further alternative method for repairing a crack 512 in a casting die block 510 is shown. Similar features in Figures 5A to 5C to those in Figures IA to 1C have been given the same reference number, but are prefixed with the numeral 5.

The method for repairing the crack 512 shown in Figures 5A to 5C, is of particular application where it is desirable to repair a conduit 511 that is a blind hole in die block 510. This may be the situation for example of a crack in a cooling fountain of a casting die.

The first step of this alternative method, as shown in Figure 5B, involves casting the alloy 520 into the blind hole 511 which expands upon solidification, thus forming a tight seal across the outer wall of the conduit 511. The second step is machining, typically by drilling, down the centre of the cast metal 520, sufficiently deep to leave an adequate thickness of metal between the bottom of the drilled hole 530 and the bottom of the blind hole 511 in the die, Figure 5C, the drilled hole being of such a diameter that an adequate skin of metal is left on the wall of the blind hole to substantially prevent any leakage of fluid from the new drilled conduit as well as loss of thermal efficiency.

Referring now to Figures 1OA and 1OB, a further alternative method for repairing a crack 712 in a die block 710 is shown. Similar features in Figures 1OA and 1OB to those in Figures IA to 1C have been given the same reference number, but are prefixed with the numeral 7.

As with the method for repairing a crack shown in Figures 2A and 2B, a tubular insert 722 is positioned inside the conduit 711 of the die block 710 and an alloy 720 is cast between the outer wall of the tube insert 722 and the inner

wall of the conduit 711. The tubular insert 722 has projections 724 on its outer wall which enable the tubular insert 722 to be centred within the conduit 711. For this purpose, the projections are arranged around the outer wall of the tubular insert 722. The projections are also laterally spaced apart along the length of the tubular insert 722 so as to not provide any significant obstruction to the casting of the alloy 720 between the tubular insert 722 and the die block 710. The projections 724 may be deposits of weld or a brazing metal or solder, or small buttons attached by welding, brazing, soldering or by any other suitable fixing method. Alternatively, the projections 724 may be formed by machining away a layer of metal to leave free standing studs, spigots or stand offs.

Referring now to Figures HA and HB, yet another alternative method for repairing a crack 812 in a casting die block 810 is shown. Similar features in Figures HA and HB to those in Figures 2A and 2B have been given the same reference number, but are prefixed with the numeral 8 instead of 1.

The method for repairing the crack 812 shown in Figures HA and HB, is of particular application in repairing a crack 812 which is located near to a join between two conduits 8HA and 8HB in the die block 810. In the

Figures of HA and HB, the conduits 8HA and 8HB are shown joined at right-angles to one another, however, the method could equally apply to conduits which did not join at right- angles . In a first step of this method, a first tubular insert 822A for one of the conduits 8HA is formed with a taper threaded end 825. A second tubular insert 822B for the other conduit 811B is drilled and tapped to produce an aperture of appropriate size and orientation to receive the taper threaded end 825 of the first tubular insert 822A. In a further step of this alternative method, the second tubular insert 822B is inserted into its respective conduit 811B and

the first tubular insert 822A is inserted into its respective conduit 8HA such that the taper threaded end 825 of the first tubular insert 822A is received in the aperture of the second tubular insert 822B. Alloy 820 is subsequently cast between the outer walls of the tubular inserts 822A, 822B and the inner wall of the conduits 811A, 811B.

It is noted that one of the tubular inserts (in this case the second tubular insert 822B) may comprise projections 824 on its outer surface for centring the tubular insert 822B in the conduit 811B. In this embodiment, the first tubular insert 822A may be centred within its conduit 8HA by location of its taper threaded end 825 within the aperture of the second tubular insert 822B.

Referring now to Figures 12A and 12B, another alternative method for repairing a crack 912 in a casting die block 910 is shown. The method for repairing the crack 912 shown in Figures 12A and 12B is of particular application where it is desirable to repair a conduit 911 that is a blind hole in the die block 910, similar to the method shown in Figures 4A and 4B. Similar features in Figures 12A and 12B to those in Figures 4A and 4B have been given the same reference number, but are prefixed with the numeral 9 instead of 4.

The first step of this alternative method involves inserting a tubular insert 922 with a sealed end into the blind conduit 911. The sealed end is provided by a plug 926 which is permanently joined to the end of the tubular insert 922 which is inserted into the blind conduit 911. The plug 926 comprises a number of projections 924 projecting from its outer walls. The projections 924 are located on the sides of the plug 926 to centre the plug 926 (and the tubular insert 922) within the conduit 911 and also on the end wall of the plug 926 to space the plug 926 away from the blind end of the conduit 911. Although the plug 926 is shown in Figures 12A and 12B as having three side projections and one end projection, it may comprise more or less projections as required.

The plug 926, internally, also comprises a space 927 that minimizes the distance between the cooling or heating fluid and the working face of the die 910. The contour of this space 927 generally fits with the contour of the blind conduit 911. The space 927 improves the thermal efficiency of the repaired conduit 911 because if the fluid is too far back from the die face, then the cooling or heating rate may be substantially reduced.

Referring now to Figures 13A and 13B, a further alternative method for repairing a crack in a casting die block is shown, which is of particular application when the crack is located near to a join between two conduits, similar to the method described with respect to Figures HA and HB. Similar features in Figures 13A and 13B to those in Figures HA and HB have been given the same reference number, but are prefixed with the number 10 instead of 8.

The first step of this alternative method, as shown in Figure 13A, involves inserting a shaped plug 1027 into one end of a first tubular insert 1022A (which is eventually to be inserted into one of the conduits) . The distal end of the shaped plug 1027 is shaped to follow the contour of the outer surface of a second tubular insert 1022B so that it may snugly abut the tubular insert 1022B (which is to be inserted into the second conduit) . In a second step in this alternative method, portions of the surfaces of the plug 1027 and the tubular inserts 1022A,1022B in the area at and around which a join is to be formed between the tubular inserts 1022A,1022B are coated with a suitable soldering flux, such as a zinc- chloride flux for example prior to being inserted into the conduits. The tubular inserts 1022A and 1022B are subsequently positioned within the conduits of the die block and a further alloy, similar to the alloy in any of the methods described above is cast between the outer wall of the tubular inserts 1022A,1022B and the inner walls of the conduits. The alloy cast around the tubular inserts 1022A, 1022B forms a metallurgical bond to the inserts where the

inserts have been coated with the soldering flux. The inserts 1022A, 1022B are thus sealed with a metal layer that is metallurgically bonded to the inserts 1022A,1022B, and hence reduces the possibility of any heating or cooling fluid flowing through the inserts 1022A,1022B during operation of the casting die from leaking. Once the tubular inserts 1022A,1022B have been joined together, the plug 1027 and the outer wall of the second tubular insert 1022B are drilled through to fluidly connect the inserts 1022A,1022B together. The methods described above can be used to repair other imperfections in the casting die block 10 such as micro cracks, fissures, indentations, fractures and chips. The imperfections which can be repaired by the method according to the present invention may also be located in positions other than in the fluid conduits of casting dies.

The methods described above can be used to preempt a failure such as the formation of a crack in the die block 10. If it is thought that a crack is highly likely to form in the die block 10 during operation of the die, the methods described above can be applied to the die block 10 before it is used so that if a failure occurs, a fluid leak is avoided and the operation of the die is not compromised. The methods described above may also be used to improve the heat transfer path of any article in which a cold body (such as a cooling mechanism) is separated from a hot body (such as a heat load) by an air gap by filling the air gap with alloy. Example articles are injection moulding dies or electronic devices such as laptop computers which have heat pipes as separate self-contained units located in a heat sink to conduct heat away from the sink. The methods described above may be employed to fill the gap between the heat pipes and the heat sink, thus improving the rate of heat transfer from the sink to the heat pipes .

The repair methods of the present invention were

developed using a cylindrical test apparatus as shown in Figure 6a. Dimensions in millimetres are shown in Figure 6b.

The test apparatus 610 is a 100mm diameter x 120mm high cylinder of mild steel with 4 x 12.7mm diameter drilled through holes 611, each of which has its centre 20mm from the circumference of the cylinder. The apparatus 610 was electrically heated with 2400W elements 630 controlled by a variable current power supply. The through holes 611 were threaded at both ends to allow plumbing connections 640 for inflow and outflow of water. Each hole 611 simulated a conduit in a die block.

The amount of heat transferred from the steel block 610 to the cooling fluid was successfully used as a measure of cooling capacity during the testing program. This Cooling Power was determined by thermocouples measuring the temperature of inflowing cooling water at the base and the out flowing water at the top of the apparatus 610, and by measuring the water flow rate. Cooling water was pumped from a Im ³ reservoir of water at room temperature .

Cooling power was calculated at thermal equilibrium using the formula:

P = Cooling Power m = mass flow rate of coolant Cp = specific heat of coolant T _out = temperature of coolant at the outlet Ti _n = temperature of coolant at the inlet

A conduit 611, as drilled and with no visible rust over the interior surface, was tested to establish a standard cooling power against which repair techniques could be compared. The heat input and the cooling water flow rates were maintained at 1500W and 0.52/min for the

duration of the test. Figure 7 shows the logged data obtained from measuring the increase in water temperature as it flowed through the heated block 610. Prom the recorded data, thermal equilibrium was achieved about 30mins after heating began and the cooling water temperature increased from 17°C to 59°C, a δT of 42°C.

EXAMPLE 1

Conduit lined with Commercial Bismuth Tin alloy A commercially available bismuth-tin alloy was cast into a conduit 611 in the test block 610 by pouring molten alloy into the preheated block (200°C) . After the alloy solidified, a flow path was re-established by drilling through the cast-in metal using a 10mm diameter bit, such as described in Figures Ia, Ib and Ic. A cooling water flow circuit and electrical heating was set up as for the control condition above, and inflow and outflow water temperatures measured to determine the cooling power of this configuration. Tests were performed with a heating rate of 1500W and a water flow rate of 0.52/m. Figure 8 shows the water temperature variation for this test . It can be seen that the time to reach thermal equilibrium was approximately 60mins and the water temperature increased from 14°C to 55 ⁰ C, a δT of 41°C.

EXAMPLE 2

Conduit repaired with Bismuth Zinc alloy and a Copper sleeve

Repairing a damaged conduit may also be performed by placing a metal sleeve into the conduit to act as a barrier between the damage and the cooling fluid, as described in Figures 3a and 3c. The space between the sleeve and the die wall is filled with expanding metal to improve heat transfer. This technique was tested using a bismuth zinc alloy, formulated in the laboratory to achieve a higher melting temperature than the bismuth tin alloy used in Example 1, as the low melting temperature of

this alloy could restrict its use. One end of the test block 610 conduit 611 was plugged using a screw-in fitting (similar to 223) . Granulated bismuth zinc alloy was placed into the plugged conduit in the test block and a copper pipe, again with one end plugged, was placed in the same conduit , above the granules . The apparatus was then preheated in an oven to 300°C. Once the block had reached this temperature, it was removed from the oven and the copper sleeve (similar to 222) forced into the molten alloy until it was seen to penetrate up to the top of the block, around the sleeve. After the block 610 was cooled and the alloy solidified, the plug at the end of the conduit was removed and the plug in the copper sleeve drilled through to restore a flow path to arrive at a configuration similar to Figure 2b. Water cooling and electrical heating was then applied to the test block as for the control condition and inflow and outflow water temperature measurements recorded. Tests were performed with a heating rate of 1500W and a water flow rate of 0.481/m. Figure 9 shows the water temperature increase as it passed through the repaired test block. Again, thermal equilibrium took approximately 60mins to establish and the temperature of the water exiting the block increased from 11°C to 52°C, a δT of 41°C. Thermocouples were also placed at various locations around and below the surface of the block 610 to monitor temperature variations during tests . These temperatures generally showed a similar variation with time as shown by the outflow water temperatures, ie. after an initial increase, the temperature reached an equilibrium value, at approximately the same time as the outflow water reached equilibrium.

The calculated cooling power of the repair techniques is shown in Table 1.

Table 1

Using this same test apparatus 610, common repair techniques used by industry practitioners were investigated. Meticulous application of these techniques in the laboratory could only produce cooling power performances to a maximum of 60% of the as drilled conduit, compared to around 95% for the preferred embodiment .

EXAMPLE 3

Intersecting conduits repaired with Bismuth Zinc alloy and Brass sleeves.

A conduit was repaired where the damaged conduit was one of three intersecting conduits forming a ^λ U' shaped circuit in a rectangular H13 tool steel block,

300x400x120mm in dimension. The steel block was designed with a split line so that it can be opened in the laboratory to examine the condition of test repair techniques. The ^V U' shaped circuit consisted of two vertical conduits 14mm in diameter, 240mm long and 200mm apart that intersected a horizontal conduit of the same diameter and 350mm long. The circuit enabled water to flow down one vertical conduit into and along the horizontal conduit and up the other vertical conduit to exit the block. All conduits were accessible from one end of their lengths .

Three brass sleeves were prepared for use in the repair; two sleeves of 12.7mm outer diameter, 1.1mm wall thickness and 300mm long with a contoured plug brazed into one end (similar to the tubular insert 1022A and the plug 1027 in Figures 13A and 13B) _/ one sleeve 330mm long of the same diameter and wall thickness plugged at both ends (similar to the tubular insert 1022B in Figures 13A and

13B) .

The steel block was placed in an oven and heated until it reached a temperature of 400 ⁰ C, after which it was removed from the oven. The two 300mm sleeves were placed on top of the block to preheat them. The 330mm long brass sleeve was coated (cold) with a commercial Zinc Chloride soldering flux around its outer surface, placed into the horizontal conduit of the steel block and the open end of the conduit sealed with a cover plate. Concurrently, approximately 30Og of a Bismuth-Zinc alloy (consisting of 7% by weight zinc with the balance being bismuth except for incidental impurities) was heated to 300°C where it was fully molten. After the horizontal conduit had been sealed as above, this molten alloy was poured into one of the vertical conduits of the steel block. Each of the two 300mm long sleeves were then coated on their outer surfaces with the same commercial zinc chloride soldering flux at their plugged end, extending up, along the length of the sleeve approximately 50mm away from the plug. One of each of these sleeves were then placed into each of the vertical conduits, (plugged end first) and rotated until the contoured end mated with the surface of the horizontal sleeve (similar to Figure 13A) . Each vertical sleeve protruded from the block by about 60mm so that a weight could be placed onto the sleeve to stop it floating in the molten alloy.

When the steel block had cooled to room temperature, the plugged ends of the vertical sleeves were drilled through with a 9mm diameter drill bit to connect the vertical and horizontal flow paths together.

The results of metallurgical testing of the repaired steel block showed that fluxing the surfaces of the sleeves facilitated a metallurgical bond to form between the sleeve and the alloy, thus increasing the strength of the joint at the intersecting conduits.

EXAMPLE 4

Intersecting conduits repaired with Bismuth-Zinc-Silver alloy and Brass sleeves.

A cooling circuit formed by three intersecting conduits in a steel block as described in Example 3 was repaired using a Bismuth-Zinc-Silver alloy consisting of 2.7% by weight zinc, 2.5% by weight silver and the balance being bismuth except for incidental impurities . Repair of the conduits occurred in accordance with the methodology of Example 3 with the three brass sleeves being plugged and arranged in the steel block as described in Example 3 and the Bismuth- Zinc-Silver alloy being melted and poured into one of the vertical conduits of the steel block.

In the preceding description of the invention and in the claims which follow, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, ie. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.