GB1600384A

GB1600384A - Submarine telecommunications cables and methods of treating such cables

Info

Publication number: GB1600384A
Application number: GB19529/77A
Authority: GB
Original assignee: Post Office
Current assignee: Post Office
Priority date: 1977-05-10
Filing date: 1977-05-10
Publication date: 1981-10-14
Also published as: FR2390815B1; JPS5416691A; FR2390815A1

Description

(54) IMPROVEMENTS IN OR RELATING TO SUBMARINE TELECOMMUNICATIONS CABLES AND METHODS OF TREATING SUCH CABLES (71) We, THE POST OFFICE, a British body corporate established by statute, of 23 Howland Street, London, W1P 6HQ, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed to be particularly described in and by the following statement: This invention relates to improvements in or relating to submarine telecommunications cables and methods of treating such cables.

It has been found that the change in dielectric loss with frequency of a submarine cable having a polyethylene dielectric may increase over a period of time such as a few years; this effect is particularly noticeable at high frequencies, for example 30 MHz. A small variation in dielectric loss of a submarine cable over the frequency range of the signals it carries may be compensated by shore-controlled adjustable equalisers; since these equalisers are required to provide varying compensation as the dielectric loss in the cable varies with frequency, such equalisers represent an elaborate and expensive solution to the problem.

Also, as the change in dielectric loss with frequency increases so the compensation by the equalisers must become more elaborate. A change in loss angle (defined below) in a cable of more than 3rad at 30 MHz, which it is believed is quite likely to occur after a cable has been in use for a substantial period of time, is at best extremely difficult to compensate with equalisers.

It will be understood that the "loss angle" of a dielectric is the angle whose tangent is the ratio of the real part of the admittance of the dielectric to the imaginary part of the admittance of the dielectric.

It is an object of the invention to provide a method of reducing the increase in dielectric loss in a submarine telecommunications cable upon prolonged exposure to water or its vapour.

According to the invention there is provided a method of manufacturing a submarine telecommunications coaxial cable in which an outer conductor surrounds a dielectric which surrounds an inner conductor, the method including reducing the increase in dielectric loss upon prolonged exposure to water or its vapour of the cable by the step of: maintaining at least the outer region of the dielectric in the cable at a selected temperature lying in the range of from 40"C to 90"C for a selected length of time, the time being sufficiently short and the temperature being sufficiently low that the performance of the cable is not impaired or any impairment is kept within acceptable tolerances.

The dielectric may be polyethylene. Alternatively the dielectric may be polypropylene or a copolymer of ethylene.

The heat treatment may be carried out in a batch process; a length of cable a few miles long may be treated as one batch. In this case it may be convenient to treat the cable at a low temperature and for a long period of time: the treatment temperature preferably lies in the range of from 50"C to 70"C and the treatment time in the range of from 1/4 hour to 200 hours. For example a heat treatment for 6 hours at 60"C has been shown to give satisfactory results. The heat treatment may be carried out before the outer conductor and sheath are added around the cable alternatively, the cable core may be dried, the outer conductor and sheath added and then the heat treatment carried out.

It will be appreciated that as the temperature of heat treatment is reduced so the necessary treatment time is increased.

Alternatively the treatment may be carried out as part of a continuous cable production process. In this case it is preferably to treat the cable for only a short period of time and therefore a higher treatment temperature should be used: the treatment temperature may lie in the range of from 50"C to 90"C and the treatment time in the range of from 15 minutes to 5 hours. For example heat treatment at 600C for times in the range of from 1 to 4 hours have been shown to give good results. The treatment may be carried out before addition of an outer conductor or a sheath around the cable.

A heat treatment at 400C for a time in the range of 100 hours to 200 hours has also been found to give similar but less effective results.

It will be appreciated that as the temperature of heat treatment is increased so the necessary treatment time is reduced.

According to another aspect of the invention there is provided a method of manufacturing a submarine telecommunications coaxial cable in which an outer conductor surrounds a dielectric which surrounds an inner conductor, the method including reducing the increase in dielectric loss upon prolonged exposure to water or its vapour of the cable by the step of: maintaining at least the outer region of the dielectric in the cable at a selected temperature lying in the range of from 40"C to 250"C for a selected length of time, the time being sufficiently short for the selected temperature that the performance of the cable is not impaired or any impairment is kept within acceptable tolerances.

As already stated the heat treatment may be carried out as a continuous process. One way of effecting the treatment is to treat the cable after adding the outer conductor. The heat treatment may be carried out by prolonging the step of sheathing the cable and may comprise covering the outer conductor with a sheath of molten material and retaining the cable at a selected elevated temperature for a selected length of time before the cable is cooled to room temperature. The molten material may be polyethylene.

The selected length of time may be more than twenty seconds. As already stated as the selected temperature of heat treatment is increased so the necessary treatment time is reduced. The selected temperature may be at least 1500C. In one heat treatment which has been found to be effective the selected temperature is approximately 1700C. In this particular heat treatment the selected length of time is approximately one minute.

According to another aspect of the invention there is provided a submarine telecommunications coaxial cable manufactured by a method as defined above.

By way of example only methods embodying the invention of reducing the increase in dielectric loss upon exposure to water or its vapour of a submarine telecommunications cable will now be described with reference to the accompanying drawings of which: Figure 1 is a partly schematic side view of an end of a submarine telecommunications cable cut away to reveal its constituent parts; Figure 2 is a graph illustrating the estimated effect of exposure to sea water on the attenuation of a conventional submarine cable; Figures 3A to 3G are graphs illustrating the effect of various heat treatments on the change in dielectric loss of a polyethylene dielectric upon exposure to high humidity; Figures 4A and 4B are diagrams showing respectively a simplified representation of a cube of wet dielectric containing a water droplet and an equivalent circuit therefor; and Figures 5A and 5B are block diagrams showing respective methods of manufacturing a submarine telecommunications cable.

The cable shown in Figure 1 is a submarine coaxial cable of the armourless type. The cable has a strength member 1 which may be of stranded construction having lays of opposite hand, contained within an inner conductor 2 which is spaced by a polyethylene dielectric 3 from an outer conductor 4. An outer protective sheath 5 surrounds the outer conductor 4.

These components are conventional and need not be described further; it will be appreciated, however, that the form of cable shown in the drawing is not essential and the components of the cable may be modified in any appropriate manner; for example, the cable may be armoured.

In the manufacture of the cable the dielectric 3 is extruded around the inner conductor 2 and the strength member 1 and the cable core so formed is passed through a series of water troughs maintained at temperatures up to 105"C. The first water trough into which the cable passes is the warmest, the subsequent troughs being maintained at progressively lower temperatures so that the cable is progressively cooled as it passes through the troughs. The cable takes about 90 minutes to pass through the series of troughs. The outer conductor 4 and sheath 5 are fixed around the cable at a later stage of manufacture.

During cooling of the dielectric 3 around the inner conductor 2, substantial hoop stresses are produced in the dielectric and these hoop stresses provide the inter-layer shear strength between the dielectric and the inner conductor.

During the passage of the cable core through the water troughs the outer region of the dielectric 3 becomes impregnated with sub-microscopic water droplets which could be as small as one tenth of a micron in diameter. Also, we have found that, in a cable core with a polyethylene dielectric having an outer diameter of 1.7 inches and an inner diameter of 0.8 inches there is substantial penetration of water only to a depth of about 0.15 inches into the dielectric. It should be understood that this "substantial penetration" of water represents concentrations of water of about 100 p.p.m. Nonetheless these water droplets which represent local regions of high permittivity and much greater electrical conductivity than the polyethylene can have a considerable effect on the loss in the cable at frequencies in the megahertz range.

Water is also absorbed by the dielectric in the vapour phase as distributed molecules; however the vapour phase contributes negligible dielectric loss in the megahertz frequency range. Water in this form is normally found in concentrations of about 10 p.p.m.

In the past, the cable core has been dried out at room temperature after the extrusion process and in this case the dielectric loss of the core reverts to a value similar to that of a polyethylene which has never been wetted. However, when such a dried out cable is exposed to water or its vapour reabsorption of water can occur, for example, under sea-bed pressure and temperature conditions. This is a particular problem when the cable is not provided with a seal at junctions with repeaters; however even in a sealed cable water permeates through the outer sheath of the cable over a prolonged period of time.

Figure 2 illustrates graphically the increase in dielectric loss of a cable upon exposure to sea water. The graph shows the approximate estimated increase in attenuation (measured in d B) of a 3400 Nm cable, the attenuation being plotted against frequency. The estimate is derived from laboratory experiments in which samples of dielectric are cooled from the molten state in a water bath at 900C for 30 minutes. The estimated attenuation relates to the cable of Figure 1 having a polyethylene dielectric with an inner diameter of 0.8 inches and an outer diameter of 1.7 inches. Three curves, referenced I, II and III are shown in Figure 2. Curves I, II and III are the approximate estimates for a cable exposed to sea water.The material representing the outer layer of the core was exposed to high humidity for 47 hours, 119 hours and 15 days respectively. In a cable the effect of exposure will proceed at a considerably slower rate due to the construction of the cable the longer diffusion path and the effects of reduced temperature etc. However it is possible over a period of many years that an increase in attenuation of over 70 d B could occur at 30 MHz in a transatlantic cable.

It should be noted that as the time of exposure of the cable increases so the variation of attenuation with frequency becomes progressively more non-linear. This variation of attenuation with frequency means that a great many equalisers must be used to compensate for the varying attenuation in the cable so that the cable can operate satisfactorily over its full frequency range.

We have found that by a suitable heat treatment of the cable core after the extrusion process not only can the cable be dried out but its capacity to reabsorb water can be greatly reduced and the increase in dielectric loss of the cable upon exposure to water can be limited. The heat treatment must be carried out at a selected temperature and for a minimum specified time.

The production of microdroplets in the dielectric during passage through the water troughs is thought to be nucleated by chemical groups which are attractive to water.

Such groups may be catalyst residues or anti-oxidation additives or polar groups. Even if the microdroplet evaporates a void containing the chemical group is left. Provided that the chemical group remains, the void may be refilled. When the material is heated it is believed that the groups disperse from their sites, the movement of the groups being greatly facilitated by the increased temperature. Once the groups are dispersed the void site becomes less hydrophylic and will not refill to the same extent; the effect of such water droplets on the change in dielectric loss will also be less than in the previous case.

The temperature of the heat treatment should be sufficiently low or the time sufficiently short not to seriously damage any parts of the cable core. The two most likely forms of such damage which may impair the performances of the cable are deterioration of the dielectric and reduction in the inter-layer shear strength between the dielectric 3 and the conductor 2 as a result of the relief of the hoop stresses in the dielectric.

In order to illustrate the effect of treating the dielectric at temperatures above room temperature, various examples of tests carried out on samples of polyethylene dielectric will now be described with reference to Figures 3A to 3G.

In all the tests a polyethylene plaque, while molten was immersed in water at a temperature of 900C and soaked for 30 minutes. This process roughly simulates the process which the dielectric in a cable would undergo as the cable core is extruded and cooled in a series of water troughs (the first of which is at 90"C). The plaque was then conditioned by heat treatments at different temperatures and for different times according to the test.

After drying the loss angle of the polyethylene was measured. The plaque was than exposed either to salt water or to moist air of about 98% relative humidity, this exposure replacing the practical case where the cable is exposed to sea-water. After various times of exposure the loss angle was again measured at various frequencies up to 30 MHz and the increase in loss angle (i.e. the excess loss angle) at the various frequencies calculated.

The graphs shown in Figures 3A to 3G show increase in loss angle plotted against frequency, the increase in loss angle being measured in microradians and the frequency in megahertz on a logarithmic scale. It should be noted that the scale for the increase in loss angle is not the same in all the Figures.

The table (Table I) below shows for each of the tests the conditioning temperature and time of each test, the time for which the plaque was exposed to the moist air or salt water before the loss angle measurements were made (exposure time) and the change in loss angle in microradians due to the exposure at frequencies of 1, 3, 10 and 30 MHz.

TABLE I Conditioning Exposure Time Loss angle change (prad) Test No. Time Temp 1 Mhz 3 Mhz 10 Mhz 30 Mhz 1 21 hr. 21"C 185 hr. 16.5 44 66.5 38.5 2 170 hr. 21"C 185 hr. 15 31.5 45.5 28 3 98 hr. 40"C 185 hr. 2.5 4.6 8.0 6.2 4 150 hr. 40"C 185 hr. 3.6 2.8 3.9 3.2 5 4 hr. 60"C 185 hr. 2.1 1.1 1.1 1.8 6 24 hr. 60"C 21 days 4.9 4.0 4.9 4.0 7 24 hr. 60"C 133 days 7.1 5.8 5.5 5.2 8 24 hr. 60"C 251 days 10.3 7.8 7.8 6.9 9 24 hr. 60"C 366 days 7.7 6.6 7.6 8.0 10 24 hr. 60"C 461 days 8.9 9.0 9.7 11.2 11 6 hr. 60 C. 19 days 3.1 2.9 2.4 2.5 12 6 hr. 600C 42 days 4.0 3.9 4.8 5.0 13 6 hr. 60"C 237 days 5.7 5.1 6.6 6.6 14 6 hr. 60"C 309 days 8.5 8.1 9.1 8.6 15 6 hr. 60"C 434 days 7.5 6.6 6.7 6.0 16 20 min. 70"C 3 days 2.5 - 1.0 0.7 17 20 min. 70"C 14 days 3.7 - 2.1 1.3 18 20 min. 70"C 66 days 5.0 3.8 3.6 3.6 19 20 min. 70"C 131 days 9.3 7.5 7.4 8.8 20 20 min. 70"C 251 days 6.5 5.1 5.1 5.5 21 17 days 14.9 27.1 34.0 18.7 No conditioning 22 1 min. 1700C 17 days 3.9 1.7 1.0 1.4 23 34 days 4.7 11 18 24 No Conditioning salt sol 24 24 hr. 65"C 35 days 1.3 1.5 1.8 3.0 salt sol The tests are numbered sequentially and the result of each test is illustrated by a respective curve in Figures 3A to 3G. In these Figures the curves are labelled T1, T2, T3....T24 and correspond to tests 1, 2, 3.... 24 respectively.

In tests T1 to T22 the plaque was exposed to moist air. In tests T23 and T24 the plaque was exposed to 2.8% salt solution.

In Figure 3A the encircled numbers on curves T1, T3 and T5 show the water content in the polyethylene after the exposure to high humidity. It will be seen that there is a clear correlation between water content and increase in dielectric loss angle. It should be understood that water will be present in the vapour phase at concentrations of about 10 p.p.m. and this concentration therefore represents a "dry" cable.

It will be clear from Figure 3A that a very substantial limitation of the increase in dielectric loss of the polyethylene upon exposure to water is achieved as a result of heating the dielectric above room temperature for a period of time.

For a transatlantic cable carrying frequencies of 30 MHz an increase of loss angle of less than 1 urad may be tolerated. Since most of the water absorbed is in the outer 0.15 inch of the dielectric it is this outer core which contributes most to the increase in loss angle. An increase in loss angle of 10 urnd in the outer region of the core results in an overall increase of only about 1 urad in the loss angle of the whole dielectric so that the heat treatments of tests T3 to T20, T22 and T24 are all satisfactory although the treatments at 600C and higher temperatures seem to give better results than treatment at 400C.

As can be seen from the results of the test T24 the exposure to salt water has a similar effect to the exposure to moist air.

The results of test T22 show that a heat treatment at quite a high temperature when applied for a short period of time is effective.

Although most of the tests were conducted on sample plaques of dielectric, in tests T23 and T24 the heat treatments were carried out on a sample cable. It was found that the results obtained from heat treatments on plaques and on cables correlated quite closely.

As the curves show the increase in loss angle is frequency dependent for samples that have not been adequately heat treated. Although the curves for the inadequately treated samples show the increase in loss angle falling with increasing frequency this is due to an increase in the imaginary part of the admittance of the polyethylene as the frequency increases rather than a decrease in the real part of the admittance. In fact the real part of the admittance of the dielectric will increase with increasing frequency. The variation of loss angle with frequency in a medium containing materials of different resistivity and permittivity is known as the Maxwell-Magner effect and a simple explanation of the effect will now be given.

The Maxwell-Wagner effect relates to the loss process that occurs in a medium, containing materials of different resistivity and dielectric permittivity. In its simplest form it deals with the case of two slabs of dielectric placed in series. Electrically this is equivalent to two series capacitors, each bypassed by a resistor. Such simple systems are rarely found in nature, but it is possible to adapt the treatment to deal with the case of a random distribution of small volumes of material.

Polyethylene containing droplets of water can be regarded as such a system. Water is slightly conductive and has a much higher permittivity than polyethylene.

The treatment given here with reference to Figures 4A and 4B is very simplified and aims only to demonstrate the form of the frequency dependence of the loss angle. Let us consider (Figure 4A) very low concentrations of water and look at one small cubic region (side a) of permittivity and resistivity E2 and p respectively, contained in a larger cube side A) of non conductive material of permittivity E.

For simplicity the field distortions that would occur are ignored and it is assumed that the field lines are as straight as if the small cubic region was not present.

The cube of dielectric can be broken into a first homogeneous region composed of regions B1 and B2 (Figure 4A) and a second compound region C. Since field distortions are ignored these regions may be considered in an equivalent circuit as being in parallel.

Thus the first region (B1, B2) which is considered to be non-conductive has a capacitance

The second region (C) has a capacitance

due to the polyethylene. in series with a capacitance C = e0e2a due to the water droplet; however the water droplet also has a resistance R = p/a, where p is the resistivity of the water.

Thus the cube of dielectric can be represented by the equivalent circuit shown in Figure 4B, namely a capacitance C0 in parallel with a series combination of a capacitance C1 and a capacitance C in parallel with a resistance R.

It can then be shown that the complex admittance Y of the cube is given by:

where k = 1 + C1/C and T = RC One may define the loss in terms of the loss angle 5 (as hereinbefore defined). Then, assuming a is much less than A it can be shown that:

If the field distortions are considered then the loss will be larger due to the field concentrations at the conductive region. According to one calculation on this basis the loss angle is given by:

Where v is the volume fraction of conductive material, namely;i;3. This equation was quoted by R.J. Meakins & J.S. Dryden in the Proceedings of the Physical Society of London, Volume 70, page 427, 1957; the equation is calculated for low concentrations of spherical regions.

On the basis of both these results one would expect the shape of the change in loss angle against frequency curve to be of the general shape shown for example in curve T1 (Figure 3A). Thus the experimental results and theory are in accord.

The heat treatment of the cable during manufacture may be applied to the dielectric in various ways. In one case illustrated in Figure 5A, the cable core is treated as a stage in continuous production of cable. Prior to addition of the outer conductor and sheathing, the cable core is heat treated using infra red heating; the infra red heater must be arranged to heat a length of cable such that at the speed at which the cable emerges from the water trough any portion of the cable is heated for the desired period of time. Since the cable core is typically produced at a rate of about 10 feet per minute it is desirable that the cable be heated for a relatively short period of time so that only a relatively short length of cable need be heated. Thus, a relatively high temperature of treatment is desirable.As an alternative to infra red heating the cable core may be heated dielectrically or by other appropriate means.

In another method illustrated in Figure SB (in solid outline) the heat treatment is carried out as a continuous process during sheathing. The cable core is first manufactured in a conventional manner and a copper tape is wound round the cable core to form the outer conductor. An outer polyethylene sheath is then extruded around the core. In known methods of manufacture the cable is passed from the extruder directly into a cooling trough at room temperature, cooling and solidifying the molten polyethylene sheath. In the method embodying the present invention, however, the cable is not passed into the cooling trough until about a minute after it has passed out of the cooling trough.During this minute the sheath material is at a temperature of a little above about 1700 and is in good thermal contact with the core to which it applies the heat treatment. As can be seen from the results of test T22 above. a heat treatment of 1700C for 1 minute has been found to give good results.

The heat treatment is sufficiently rapid that despite the relatively high temperature of treatment the quality of the cable is not degraded. In the example just described the temperature of treatment is about 1700C, but it is thought that treatment temperatures up to 250"C would be satisfactory. In the particular example just described the sheath is of polyethylene but if sheaths of other material were employed other temperatures of treatment might prove advantageous.

The application of the heat treatment within a period of time of the order of one minute or so, makes the treatment easy to apply as a continuous process which in turn makes it susceptible to industrial application.

In another method also of the form illustrated in Figure SA, the cable is heat treated in batches. It is normal factory practice for the cable to be collected on a drum as it emerges from the water trough and the length of cable which may, for example, be about 5 miles long, is then placed in a heated room and warm air is blown over the cable. In order to ensure uniform heating of the cable, the drum may include spacers to separate the turns of the cable. The warm air is blown from a fan located in the centre of the drum. When heating the cable core by this method it is desirable for the cable to be heated at a relatively low temperature for a relatively long time since this will reduce the effect of any differential heating of the cable and the room will be more easily maintained at the relatively low temperature.

Instead of water cooling the cable core and subsequently re-heating the core the cable core may be passed through fewer water troughs the last of which may for example be at a temperature of 70"C. The cable core is then immediately wiped to remove the excess moisture from the surface of the dielectric and a selected heat treatment applied while the cable core is still warm.

It should be understood that various other methods of heating the cable may be used. For example it may be desirable to dry the cable at room temperature before heat treating the cable since this may enable a shorter heat treatment time to be used. Also, if the cable is dried first, the heat treatment may be carried out after the outer conductor and sheath have been located around the cable core; this process is illustrated in dotted outline in Figure 5a.

Since it is only the outer region of the dielectric which contains the water it is only this outer region which must be treated. If the treatment time is sufficiently short is is possible to treat the outer region at a temperature which would damage another region of the cable were it also at that temperature; for example it is possible to treat the outer region of the dielectric at a temperature which would destroy the inter-layer shear strength between the inner surface of the dielectric and the conductor if it prevailed in that region also.

It should be understood that some small deterioration of the dielectric during the heat treatment may be tolerable since this increase in the dielectric loss in the cable may be more than offset by the reduction in dielectric loss produced by the heat treatment.

By heating the outer region of a dielectric of a submarine telecommunications cable as described above in any of tests 3 to 14, not only is the dielectric dried but the properties of the dielectric are changed and the reabsorption of water by the cable upon exposure to water or its vapour and the effect of the water on the cable loss angle is considerably reduced thus reducing the increase in dielectric loss of the cable which occurs on prolonged exposure of the cable to water or its vapour.

WHAT WE CLAIM IS: 1. A method of manufacturing a submarine telecommunications coaxial cable in which an outer conductor surrounds a dielectric which surrounds an inner conductor, the method including reducing the increase in dielectric loss upon prolonged exposure to water or its vapour of the cable by the step of: mantaining at least the outer region of the dielectric in the cable at a selected temperature lying in the range of from 40"C to 90"C for a selected length of time, the time being sufficiently short and the temperature being sufficiently low that the performance of the cable is not impaired or any impairment is kept within acceptable tolerances.

2. A method as claimed in claim 1 in which the dielectric is polyethylene, polypropylene or a copolymer of ethylene.

3. A method as claimed in claim 2 in which the dielectric is polyethylene.

4. A method as claimed in any preceding claim in which the heat treatment is carried out in a batch process.

5. A method as claimed in claim 4 in which the treatment temperature lies in the range of from 50"C to 70"C and the treatment time in the range of from 1/4 hour to 200 hours.

6. A method as claimed in claim 4 or 5 in which the heat treatment is carried out before the outer conductor and sheath are added around the cable.

7. A method as claimed in claim 4 or 5 in which the cable core is dried, the outer conductor and sheath added and then the heat treatment carried out.

8. A method as claimed in claim 1 or 2 in which the heat treatment is carried out as part of a continuous cable production process.

9. A method as claimed in claim 8 in which the treatment temperature lies in the range of from 50"C to 90"C and the treatment time lies in the range of from 15 minutes to 5 hours.

10. A method as claimed in claim 8 or 9 in which the treatment is carried out before addition of an outer conductor or a sheath around the cable.

11. A submarine telecommunications coaxial cable manufactured by a method as claimed in any preceding claim.

12. A method of manufacturing a submarine telecommunications coaxial cable in which an outer conductor surrounds a dielectric which surrounds an inner conductor, the method including reducing the increase in dielectric loss upon prolonged exposure to water or its vapour of the cable by the step of: maintaining at least the outer region of the dielectric in the cable at a selected temperature lying in the range of from 40"C to 250"C for a selected length of time the time being sufficiently short for the selected temperature that the performance of the cable is not impaired or any impairment is kept within acceptable tolerances.

13. A method as claimed in claim 12 in which the dialectric is polyethylene, polypropylene or a copolymer of ethylene.

14. A method as claimed in claim 13 in which the dielectric is polyethylene.

15. A method as claimed in any of claims 12 to 14 in which the heat treatment is carried

**WARNING** end of DESC field may overlap start of CLMS **.

Claims

**WARNING** start of CLMS field may overlap end of DESC **. Instead of water cooling the cable core and subsequently re-heating the core the cable core may be passed through fewer water troughs the last of which may for example be at a temperature of 70"C. The cable core is then immediately wiped to remove the excess moisture from the surface of the dielectric and a selected heat treatment applied while the cable core is still warm. It should be understood that various other methods of heating the cable may be used. For example it may be desirable to dry the cable at room temperature before heat treating the cable since this may enable a shorter heat treatment time to be used. Also, if the cable is dried first, the heat treatment may be carried out after the outer conductor and sheath have been located around the cable core; this process is illustrated in dotted outline in Figure 5a. Since it is only the outer region of the dielectric which contains the water it is only this outer region which must be treated. If the treatment time is sufficiently short is is possible to treat the outer region at a temperature which would damage another region of the cable were it also at that temperature; for example it is possible to treat the outer region of the dielectric at a temperature which would destroy the inter-layer shear strength between the inner surface of the dielectric and the conductor if it prevailed in that region also. It should be understood that some small deterioration of the dielectric during the heat treatment may be tolerable since this increase in the dielectric loss in the cable may be more than offset by the reduction in dielectric loss produced by the heat treatment. By heating the outer region of a dielectric of a submarine telecommunications cable as described above in any of tests 3 to 14, not only is the dielectric dried but the properties of the dielectric are changed and the reabsorption of water by the cable upon exposure to water or its vapour and the effect of the water on the cable loss angle is considerably reduced thus reducing the increase in dielectric loss of the cable which occurs on prolonged exposure of the cable to water or its vapour. WHAT WE CLAIM IS:

1. A method of manufacturing a submarine telecommunications coaxial cable in which an outer conductor surrounds a dielectric which surrounds an inner conductor, the method including reducing the increase in dielectric loss upon prolonged exposure to water or its vapour of the cable by the step of: mantaining at least the outer region of the dielectric in the cable at a selected temperature lying in the range of from 40"C to 90"C for a selected length of time, the time being sufficiently short and the temperature being sufficiently low that the performance of the cable is not impaired or any impairment is kept within acceptable tolerances.

3. A method as claimed in claim 2 in which the dielectric is polyethylene.

14. A method as claimed in claim 13 in which the dielectric is polyethylene.

out as a continuous process.

16. A method as claimed in claim 15 in which the outer conductor is added before the heat treatment is carried out.

17. A method as claimed in claim 16 in which the heat treatment comprises covering the outer conductor with a sheath of molten material and retaining the cable at an elevated temperature for a selected length of time before the cable is cooled to room temperature.

18. A method as claimed in claim 17 in which the molten material is molten polyethylene.

19. A method as claimed in any of claims 12 to 18 in which the selected length of time is more than twenty seconds.

20. A method as claimed in claim 19 in which the selected temperature is at least 1500C.

21. A method as claimed in claim 20 in which the selected temperature is approximately 1700C.

22. A method as claimed in claim 21 in which the selected length of time is approximately one minute.

23. A method as claimed in claim 12 in which the temperature and time of the heat treatment is substantially the same as the temperature and time of the heat treatment described in any one of tests T3 to T20, T22 and T24.

24. A submarine telecommunications coaxial cable manufactured by a method as claimed in any of claims 12 to 23.

25. A submarine telecommunications coaxial cable as claimed in claim 24 and substantially as herein described with reference to and as illustrated by Figure 1 of the accompanying drawings.