GB2190457A - Hydraulic cable installation system - Google Patents

Abstract

A method of installing an optical fibre or electrical cable (26) comprises installing a ductlet (25) in e.g. a water main pipeline (24) and carrying a cable (26) through the ductlet (25) using liquid pumped from one end and of similar density to the cable, the friction between the flowing liquid and the cable and the buoyancy of the cable, carrying the cable through. <IMAGE>

Description

SPECIFICATION Hydraulic cable installation system This invention relates to installing cables in pipes or ducts, particularly but not exclusively optical fibre cables in pipes or ducts containing liquid.

British Patent Specification 21 22367A discloses a method of laying a cable in a liquid pipeline and also discloses a cable design suitable for laying in a pipline.

Thistechnique has been shown to be successful but wherethe length ofthe cable becomes excessive in a pipeline in which the liquid is flowing, then significant longitudinal stress can be suffered by the cable after installation due to the frictional drag of the flowing liquid on the surface ofthe cable.

It is an object ofthe present invention to provide a method of installing a telecommunications cable in a pipeline which does orwill maintain a flowing liquid and in which the length of cable in the pipeline is substantial e.g. more than one kilometre.

According to the present invention there is provided a method of installing a telecommunications cable in a duct or pipe which will in use carry a flowing liquid, the method comprising providing within and aiong the length of the first duct or pipe a second pipe whose diameter is smallerthan thefirst one but large enough to enablethe cable to be drawn through the second pipe by means of an hydraulicfluid, and drawing the cable through the second pipe by said hydraulic fluid.

According to another aspect of the presentinven- tion there is provided a method of installing an optical fibre telecommunications link, comprising providing an optical fibre cable which contains a plurality of opticalfibreswithin a liquid-tightsheath, providing a liquid-tightductletalongthe route to be followed by the link, and carrying the cablethrough the ductlet by haudraulicfluid pumped from one end ofthe ductlet.

Preferablythesecondpipe hasa diameterof about 20mm or less and is clipped to the top or side ofthe first pipe. The first pipe may have an internal diameter ofthe order of 2.5m.

Preferablythe cable has near-neutral buoyancy in the hydraulicfluid.

Preferably the second pipe is an extruded plastics pipe of e.g. Nylon (RTM) or polyethylene.

In orderthatthe invention can be clearly understood reference will now be madeto the accompanying drawings in which:- Fig. 1 shows in cross section a first cable suitable for hydraulic insertion through a pipe according to a first embodiment of the invention; Fig. 2 shows in cross section a second cable which is an alternative to that shown in Fig. 1; Fig. 3shows diagramatically a pipe-shaft-station systefe,at a water pumping station and in which an indian installed cable according to an embodiment of ttle invention is terminated;; Fig. 4 shows schematically partofa water pipe system in which the cable of Fig. 1 or Fig. 2 has been installed, and Fig. 5 shows diagrammatically an installation system forthe cable of Figs. 1 and 2.

Referring to Fig. 1 ofthe drawings the cable there shown comprises an extruded core 1 in which are embedded a plurality of acrylate coated fibres 2 around a central member 3. This can be made in a number of ways but an advantageous way is the technique described in our co-pending patent application 8235740 (L. R. Spicer32).

Around the central core 1 is provided a Kevlar strength member 4which comprises woven threads of Kevlar and this provides significant tensile strength to the cable. Around the Kevlar strength member is extruded an outer low-density polymer jacket5 offor example polyethylene.

The overall density ofthecomposite cablewill be in the order of unity butwithin the range 0.85 to 1.15.

The overall diameter ofthe cable just described is of theorderof6mm.

As an alternative to the cable embodiment shown in Fig. 1 an alternative arrangement is possible and this is shown in Fig. 2.

Referring to Fig. 2 a central strength member comprises a central core 10 of glass reinforced polymersurrounded by a polyamide layer 11 of for example Kevlar and this is oversheathed with polyethelene layer 12. Secondary coated optical fibres 13 are stranded helically around the central strength member and this is then oversheathed with a lowdensity thermoplastic, for example polyethelene.

The interstices between the fibres a refilled with a blocking material 14 to prevent the ingress of water.

The bend performance ofthis cable will probably be superior to that ofthe cable shown in Fig. 1 because the fibres are stranded around the central member.

Theflexural modulus can beoptimised by selecting the correct proportions of glass reinforced polymer 10 to the polyamide layer 11. Howeverthis design of cable is more expensive than that shown in Fig. 1 and it has a lowerfibre countthan the Fig. 1 forthe same size of cable although it is pointed out that the same number of fibres are shown in each cable. However the outer diameter ofthe cable of Fig. 2 would be ofthe orderof6mmto 1 Omm.

Referring now to Fig. 3 there is shown part of a water main pipe system in which a cable according to Fig. 1 or Fig. 2 has been installed.

Referring to Fig. 3 a pumping station 20 houses the top end of a surge shaft 21 communicating with the junction region 22 between main water pipes 23 and 24. Valves such as 25 are provided in the pipeline to control, open and close off the flow of water as appropriate.

Pipes 23 and 24 have an internal diameter of the order of 2.5 metres and inside pipe 24 is shown a second pipe 25 having a diameter at least an order of magnitude smallerthan that ofthe main pipe 24. A schematic perspective view of a pipe incorporating the secondary pipe and the optical fibre cable is shown more clearly in Fig. 4. As shown in Fig. .4 th e seco nda ry pipe 25 is secured to the wall ofthe main water pipe 24 by means of clips 27 so that the secondary pipe 25 is secured fastwith the wall ofthe main pipe 24. This will ensure that with the flow of water along the main pipe 24 no movement ofthe cable 26 will occur since it is isolated by means ofsecon dary be 25 from the moving water in the pipe 24.

One technique for installing the cable is shown in Fig. 5. Before the pipeline 24 is filled with water it is proposed to run the secondary pipe 25 into the pipeline 24 and secure itto the walls ofthe pipeline 24 with clips such as 27. This can be achieved as the pipeline 24 is erected because it consists of spun concrete sections 2.5 metres in diameter and about 2 metres long. However it can be installed afterthe pipeline has been erected ifthe pipeline is big enough for workmen to subsequently install the secondary pipe 25 and secure ittothewalls.

Then the cable 26 is carried into the pipe 25 using a hydraulic liquid and this is shown more clearly in Fig.

5. Referring to Fig. 5 the pipeline 25 at its inlet end 25a is sealed to a housing 31 its outlet is spaced apartfrom it inlet by a distance in the range 1 Km to 10Km. The housing 31 has a variableaperture32 intheform of a cable gland through which the cable 26 is pushed by a drive unit 33 comprising a pair of tyred wheels 34 and 35 with torque monitor and control system 36. Wheels 34and35squeezethecable26and push the cable through the cable gland 32.

Thevariable aperture can be formed by a hydraulicly pressurised diaphragm whose aperture is variable in size byvarying the hydraulic pressure applied. Thus the entry of the cable through the diaphragm can be adjustedtojustfitthecable26to minimise leakage of liquid outofthe pipeline.

The housing 31 has connected to it an hydraulic37 which pumps hydraulic fluid e.g. watervia a flow meter38 intothe housing 31.Apressure gauge 39 monitors the pressure in the housing.

It is found that with the cable substantially neutrally buoyant the frinctional effect ofthe hydraulic fluid, for example water, on the cable is sufficientto draw it through pipeline 25 over distances of up to 10 kilometres. It is a particular advantage in this application which is envisaged because the main pipeline 24 will run substantially horizontally so thatthere will be little if any static induced pressure heads of liquid within the long distance travelled by the secondary pipe 25. Although of course there will be a static head where the pipe rises uptothe pumping station 20 such as is illustrated in Fig. 3.Thus it is envisaged that the pipeline shown in Fig. 4will extend over a distance of up to 10 kilometres between two pumping stations each as shown in Fig. 3 with the cable being pushed and carried from one cable gland in one pumping station and retrieved up to 10 kilometres away at the next pumping station.

Repeaters for repeating the signals in the optical fibrecablewould be installed in the pumping station.

The leading end of the cable will be protected with a watertight sock to prevent water ingress to the front end ofthe cable.

Example Cable diameter dl 6 x 10-3m Ductlet diameter d2 20 x 10-3m Liquid (water) density p 103 kg m-3 Density difference cable-liquid Ap 102 kg m 3 Coefficient of friction between cable and duct p -0.3 Liquid viscosity # 1.2x10-3 NS m-2 Liquid velocity V (ms-1) TBD Cable velocity Vc (ms-1) TBD It is expected that in order to achieve a reasonable drag force, the fluid flow will be turbulent. The liquid velocity is therefore not expected to vary significantly at different points within the ductlet, except when passing from that part ofthe ductlet containing the cableto the empty part.

The drag force per metre on the cable is: FcD = 0.5 - CFC. P.(V-VC)2. Il.dl' k (1) Where CFC is the skin friction coefficient between cable and liquid and k isthe proportion ofthe cable circumference which experiences the drag (0.5 < k < 1).

The drag force per metre which the liquid exerts on the ductlet (pipeline 25) is: F1D - 0.5 -CFD'P'V2 I) d2 (2) where CFD is the skin coefficient between ductlet and liquid.

At the flow rates likely within this system, the skin coefficients CwilI vary with Reynolds number Re roughly in accordance with the Blasius law CF - 0.079 Re-025 (3) where Re = # # V (4) i.e. the CFis a slowly varying function ofthe system parameters, and will be typically ofthe order of 0.01 (dimensionless).

The total force per metre exerted by the water in the (pipeline 25) ductlet is therefore: Ftot (liquid) = FCD + FDD (5) and this force, divided by the area of the ductlet containing the water, gives the pressure drop per metre dP = (FcD + FDD) x 4 dl Il (d22 - d12) (6) The total force on the cable is the drag force FCD plus an additionai force dueto the pressure drop x cable area Ftot.cable = FCD + IId1 dP (7) 4 dl This force, Ftot.cable, will tend to move the cable through the ductlet. It will be opposed by the frictional force between the cable and the ductlet.

Ignoring bends for the moment (see section d) this frictional force per metre is given by: Ffrict. = II d12.Ap.g. (8) 4 and the cable will move along the ductlet provided Ftot.cable > Ffrict. (9) In practise, the pressure orflowwould be adjusted so that Foot table is greater, but not very much greater than Ffrict sothatthe net force on the cable during installation is not excessive.

Simplification Equations (1) to (9) describe completely the relationships between forces, flow-rates, pressures etc.

forthe installation of a cable in a straight ductlet. They can be solved as they stand, using an iterative computer method, but it is useful to make some simplifying assumptions.

Assuming k= 1, V Vc and CFc=CFD=CF, equations (1), (2) and (6) simplifyto: dP= 0.5 . CF . # . V2 . II . (d1+d2)x 4 (10) dl II (d2+dl) (d2-di) dP = 2 CF V2 dl d2 - d1 and FcD = 0.5 CF p V2 II dl FCD= dP d2-d1 . II d1 (11) dl 4 Substituting (11) into (7) we obtain: Ftotcable = dP. II d1 d2 (12) dl 4 Using (12), (8) and (9), the criterion for movement becomes: dP .II . d1 d2 > II d1. #p . g . dl 4 4 dP > (d1/d2).#p. .g (13) dl Using the nominal numerical values quoted in section 2(a), the pressure gradient necessary for movement is 0.088 MPa/km (1 2psi/km). This corres- ponds to a flow velocity in the part ofthe ductlet containing the cable of 0.24 m/s, or a flow rate of 70 ml/s. A density mismatch of 10% has been assumed, In practise it is likely that cable and liquid densities can be matched much more closely than that.

Nevertheless, it can be seen that with the figures assumed installations over many km are achievable at quite moderate pressures.

Effect of Bends in the Duct Foracable under atension Trounding a bend of6 radians the tension change is: #T=T.(1 - e-u0) (14) This will be the only tension change due to the bend, provided that the overall flexural modulus of the cable is sufficiently low.

It can be seen in equation (14)that as the tension in the cable decreases, so does the tension change at the bend. In the limiting case, if tension at the "feed" end dropped to zero (cable slack), the frictional effect at the bend would also be zero, so that cable movementdueto flow would still take place.

The effect of bends in the duct may therefore be to slow down the rate of cable installation, but they will not stop it.

During installation there will be a flow of water out ofthe far end ofthe ductlet 25 for some considerable time before emergence ofthe cable. In some environments this waterflow could be a serious practical problem. It is, however, unlikely to be a problem for situations such as a Water Main.

It is to be expected that the flow rates and pressures actually used during installation will be significantly higherthanthose minimum quantities calculated above.

While in principle this installation technique could be applied to copper-wire cables provided they met the density requirement, these cables are not generally installed in long lengths, they can tolerate high strains (1 % ), and can be jointed relatively easily. It is therefore with optical fibre cables that the main advantages ofthis installation method becomes apparent, and its use is envisaged almost entirely with thistype of cable. It is not necessarily limited to any particular optical cable type (tight construction, loosetubeoropen channel).

For installation of long lengths the overall cable SG should be as close as possibletothatofthe impelling fluid, andthe density, orSG, mismatch should certainly not begreaterthan 10%.

Unless ductlet sizes and pumping rates are in creased to highervalues than those considered in section (2), the cable diameter will typically be 6mm (-2mm + 4mm).

Assuming a pressure drop of 0.1 Mpa/km, equation (1) gives a drag force of 9 N/km. Assuming a maximum installed length of 1 Okm, with a stationery cable and negligiblefriction, this gives a maximum force of 90 N (say 100 N). If this corresponds to 0.2% fibre strain and there is negligible strain relief in the cable, it corresponds to a cable load for 1 % strain of 500 N.

The cable flexural modulus needs to be sufficiently high forthe cableto be driven through the inlet gland, but sufficiently low to have negligible effect at any bends in the ductlet.

Claims (15)

1. A method of instal ling a telecommunications cable in a duct or pipe which will in use carry a flowing liquid, the method comprising providing within and along the length ofthefirst ductor pipe a second pipe whose diameter is smaller than the first one but large enough to enable the cable to be drawn through the second pipe by means of a hydraulic fluid and the frictional effect between the hydraulic fluid and the external surface ofthe cable, and drawing the cable through the second pipe by said hydraulic fluid.

2. A method as claimed in claim 1 wherein the secondary pipe has a diameter of about 20mm or less.

3. A method as claimed in claim 1 or claim 2, wherein the secondary pipe is clipped to the top or the side ofthe first pipe or duct.

4. A method as claimed in any preceding claim wherein the first pipe or duct has an internal diameter ofthe order of 2.5 metres.

5. A method as claimed in any preceding claim wherein the cable has a near neutral buoancy in the hydraulicfluid.

6. A method as claimed in any preceding claim wherein the second pipe comprises an extruded plastics pipe.

7. A method of installing a telecommunications in a duct or pipe which will in use carry a flowing liquid, substantially as hereinbefore described with reference to and as illustrated in accompanying drawings.

8. Acablefor installation by a method as claimed in any preceding claim, said cable comprising a plurality of optical fibres tightly held within the cable structure, a flexible tensile strength member, and an outer low-density polymer sheath.

9. A cable as claimed in claim 8wherein the tensile strength member comprises a high tensile layer of plastics threads.

10. A cable as claimed in claim 9, wherein the tensile strength member incorporates glass reinforced polymer.

11. A cable for installation ion a pipeline by hydraulic fluid friction, substantially as hereinbefore described with reference to and as illustrated in Fig. 1 of Fig. 2 ofthe accompanying drawings.

12. A method of installing an optical fibre telecommunications link, comprising providing an optical fibre cable which contains a plurality of optical fibres within a liquid-tight sheath, providing a liquid tight ductlet along the route to be followed by the link, and carrying the cable through the ductlet by hydraulic fluid pumped from one end of the ductlet.

13. A method as claimed in claim 12, wherein said fluid has a density similarto that ofthe cable.

14. Amethod as claimed in claim 13,wherein the density of the liquid is slightly greaterthan that of the cable.

15. A method as claimed in claim 12to 14wherein the liquid is pumped at a linearvelocity in the range 1 to 1 Om/sec.