Note: Descriptions are shown in the official language in which they were submitted.
12S83~
This invention relates to optical fibre
transmission lines, and in particular thouyh not
exclusively to cable structures for their installation.
Optical fibre cables carrying optical fibre
transmission lines have heretofore been installed by the
same methods as conventional metal conductor cables, those
methods usually involving pulling the cable with a pulling
rope through a previously laid cable du~t. Frequently the
cable duct already contains one or more conventional
cables at the time of installing the optical fibre cable.
Unlike the metal conductors of a conventional
cable, the optical fibres are easily damaged by tensile
stress. Such stress may, for example, propagate micro-
cracks, leading to fibre breakage in the long term. It
is, therefore, standard practice to reinforce optical
fibre cables by providing a central strength member,
usually one or more steel tension wires, about which the
optical fibres are disposed. The strength member takes
up, and thus increases the ability of the cable to
withstand, tensile stresses accompanying installation of
the cable.
Unfortunately, the central strength member
usually provides insufficient protection against local
stresses caused by pulling a further cable through the
same duct. The conventional approach of installing at the
outset optical fibre cables containing sufficiently large
numbers of optical fibres to satisfy foreseeable future
traffic demands is a way of overcoming this problem. In
consequence, first time installation of optical fibre
cables containing dozens or even hundreds of optical
fibres are currently envisaged despite the fact that to
begin with a small fraction of the installed fibres would
provide ample trafic carrying capacity. A further reason
for installing optical fibre cables of comparatively large
dimension is that the smaller the cross-section of the
cable the more prone the cable becomes to wedging in
between those cables already present in the duct.
~q~
``` i~5839(~
The first time installation of large diameter
optical fibre cables with high numbers of optical fibres,
is, however, undesirable for a variety of reasons.
Firstly, there are problems of a technical nature inherent
in such cables, such as for example the difficulty of
forming joints and of achieving the required high
strength-to-weight ratios. Secondly, there are clear
economical drawbacks in committing large resources to
install initially unused fibre capacity, particularly in
view of the comparatively recent origins of optical fibre
technology which lead one to expect continued substantial
reductions in the price and improvement in the quality of
optical fibres. Thirdly, there is the serious risk of
damaging in a single incident very large numbers of
expensive optical fibres and, finally, there is an
appreciable loss in flexibility when routing high density
optical fibre transmission lines.
A method of installing optical fibres with
pulling ropes and pull cords is described in "Sub-ducts.
The Answer to Honolulu's Growing Pains", Herman S. L. Hu
and Ronald T. Miyahara, Telephony, April 7, 1980, pp 23 to
35. The installation method described there proceeds as
follows: A section of existing 4-inch (lOOmm) duct is
rodded and thereafter between one and three individual 1-
inch t25mm) polyethylene tubes are inserted into the ductusing pulling ropes. The polyethylene tubes form sub-
ducts into which an optical fibre cable can be pulled with
the aid of a nylon pull cord which has previously been
inserted into the subduct by means of a parachute attached
to its leading end and pushed through the subduct with
compressed air.
The method just referred to does deal with some
of the problems discussed above, but only to a very
limited extent. Thus, it enables fibre capacity to be
increased in up to three stages, and separates the optical
fibre cables from those cables already in the duct,
thereby greatly reducing the likelihood of jamming, and
hence overstressing, of the optical fibre cable.
~;~S~33~(~
It is an object of the present invention to
provide an optical fibre cable structure which overcomes
or at least appreciably mitigates the majority of the
aforementioned problems.
According to one aspect of the present invention
there is provided an optical fibre cable structure
compxising, a conduit having one or more ductlets capable
of loosely accommodating an optical fibre member, and at
least one lightweight and flexible optical fibre member
inserted into an associated ductlet by propelling the
fibre member by fluid drag of a gaseous medium.
Another aspect of the invention provides an
optical fibre member comprising one or more optical fibres
contained in an outer envelope, said fibre member being
sufficiently lightweight and flexible to be propellable
along a tubular pathway by fluid drag of a gaseous medium
passing over the fibre member at a high average relative
flow velocity.
A particular aspect of the invention provides an
optical fibre cable structure of extended length
comprising: a conduit having one or more tubular ductlets
disposed therewithin, said ductlets having openings only
at either end thereof and being capable of loosely
accommodating at least one optical fibre member, and at
least one optical fibre member inserted into and through
an associated ductlet by propelling the fibre member from
one end to the other of said ductlet using fluid drag of a
gaseous medium distributed along the inserted length of
said fibre member.
It is to be understood that more than one fibre
member may be provided along a common tubular pathway.
The fibre member may, for example, comprise a
single optical fibre, protected by at least a primary
coating but preferably contained within an outer envelope.
Alternatively, the fibre member may comprise a plurality
of optical fibres contained with1n a common envelope.
The envelope may be loosely or tightly
surrounding the fibre or fibres.
~;~S839(~
- 3a -
The gaseous medium i6 chosen to be compatible
with the environment in which the invention is performed,
and in ordinary environments will be a non-hazardous gas
or gas mixture.
With the proviso about compatibility with the
environment, the gaseous medium is preferably air or
nitrogen.
The ductlets and/or the fibre members are
conveniently but not necessarily of circular cross-
section, and the fibre member is always smaller than the
ductlet in which it is installed.
In practice the ductlet internal diameter will
generally be greater, and frequently much greater than
lmm, and the external diameter of the fibre member greater
1~ than 0.5mm.
A preferred range of diameters for the ductlet
is 1 to lOmm, conveniently between 3 and 7mm, and a
preferred range of diameters for the fibre members is 1 to
S~339(~
4mm, although much larger diameters may be used provided
the fibre member is sufficiently lightweight and flexible.
The diameter of the fibre members is preferably chosen to
be greater than one tenth, and conveniently to be about
one half of the ductlet diameter or greater (and
appropriately less, of course, if more than one fibre
member is to be propelled through the same ductlet).
Insertion of a fibre membe~ by means of the
fluid drag of a gas passing over the fibre member has
several advantages over methods involving pulling an
optical fibre cable with a pull cord~
Firstly, the extra step of providing a pull cord
is eliminated.
Secondly, using the fluid drag of a gaseous
medium pxoduces a distributed pulling force on the fibre
member. This is particularly advantageous if the
installation route contains one or more bends. If, as
would be the case with a pulling cord, the pulling force
were concentrated at the leading end of the fibre member,
any deviation of the ductlet from a straight line would
greatly increase friction between the fibre member and the
internal walls of the ductlet, and only a few bends would
be sufficient to cause locking of the fibre member. The
distributed pulling force produced by the fluid drag, on
the other hand, enables bends to be negotiated fairly
easily, and the number of bends in a given installation is
no longer of much significance~
Thirdly, the fluid drag substantially reduces
overall pulling stress on the fibre member and so permits
the fibre member to be of relatively simple and cheap
construction.
Furthermore, because the fibre member is not
subjected to any substantial pulling stress during
installation, little allowance, if any, needs to be made
for subsequent relaxation.
Since the conduit can be installed without
containing any optical fibres, conventional rope pulling
12~8390
and similar techniques may be freely employed for
installing the conduitO
The capacity of a transmission line can readily
be adapted to requirements. Thus, while initially only
one or two fibre members may be sufficient to carry the
traffic, the conduit may contain a much larger number of
ductlets than are required at the time of installation,
and further fibre members may be inserbed later on as and
when needed. The conduit of the present invention is
cheap compared to the cost of the fibres, and spare
ductlets to accommodate further fibres as and when extra
capacity is required can thus be readily lncorporated
without adding more than a small fraction to overall
costs.
The present invention also permits the
installation of improved later generations of optical
fibre transmission lines. It is possible, for example, to
install at first one or more fibre members incorporating
multimode fibres, and at a later date add, or replace the
installed multimode fibre members with fibre members
incorporating monomode fibres. Installed fibre members
may conveniently be withdrawn from the ductlet, and
replacement fibre members be inserted by using the
aforesaid method of propelling by fluid drag of a gaseous
medium.
It will be appreciated that the present
invention largely avoids the risk, inherent in handling
optical fibre cables with a large number of fibres, of
accidentally damaging before or during installation in a
single event a large number of expensive optical fibres.
The present invention also enables the
installation of continuous optical fibres over several
installation lengths without joints.
Furthermore, individual fibre members routed
through the conduit can be routed, without requiring fibre
joints, into different branch conduits at junation points.
The present invention will now be explained
~2S~3390
further by way of example and with reference to the
accompanying drawings of which;
Figure 1 is a cross-section through a conduit
- suitable for implementing the invention;
Figures 2 and 3 are relatively enlarged cross-
sections through fibre members;
Figure 4 is a schematic diagram of apparatus for
inserting fibre members into ductlets by fluid drag;
Figure 5 is a schema~ic drawing of a junction
between a trunk and a branch conduit;
Figure 6 is a schematic diagram to illustrate
notation used in drag force calculations;
Figure 7 is a schematic section of a modified
drive unit; and
Figure 8 is a graph of drag force vs pressure.
Referring first to the Figure 1, there is shown
a conduit 11 incorporating six ductlets 12, one of which
contains a fibre member 14, and a core 13.
The conduit 11 is made of extruded polymer or
other suitable material, the ductlets, or bores, 12 being
formed in the conduit during its extrusion. The central
core 13 contains copper wire pairs required for testing
operations during and after installation, repeater
supervision, power supply, and the like. Alternatively,
or additionally, the core 13 may incorporate
reinforcements, for example tension wires, to take up the
tension forces during installation of the conduit.
Where required, the conduit may be surrounded by
a water barrier (not shown).
The copper wire pair for testing ca~ be omitted
from the core 13 if suitable alternative testing
facilities are available, such as, for example testing
methods using optical fibres inserted subsequently into
the conduit as described below.
Figure 2 is a cross-section through a fibre
member 21 which is in a form particularly suited for
installation by fluid drag. The fibre member 21 comprises
several optical fibres 22 lying loosely in a polymer
lZ5~339~
sheath 24. In view of the virtual absence of any pulling
stress during installation of a fibre member by fluid
drag, the fibre member 21 does not require reinforcement.
The relatively simple construction also leads to lower
production costs, as well as making the .Eibre member 21
comparatively li.ght, thereby enabling easy installation by
fluid drag.
In certai.n circumstances it may be desirable to
provide a reinforced fibre member, and Figure 3 is a
1~ cross-section through such a fibre member 31 whi.ch,
provided it is made light enough and flexi.ble enough, is
suitable for insertion by fluid drag into a ductlet 12 of
the conduit 11 of Figure 1. The fibre member 31 consists
of a plurality of optical fibres 32 arranged round a
strength member 33 and enclosed in a polymer sheath 34.
The installation of an optical transmission line
proceeds as follows:
The flexible conduit 11 is installed into an
existing duct (not shown) by conventional methods such as
~0 pulling with a pulling rope.
Because the conduit 11 does not contain any
optical fibres at this stage, the conduit 11 can be
handled in the same way as an ordi.nary cable, and no
special care needs to be taken over and above that
customary in installing conventional metal conductor
cables. If required, it is also possible at this stage,
that is before the conduit contains any optical fibres, to
pull a further condui.t through the duct to provide spare
capacity.
Furthermore, since the conduit can readily be
made of an external diameter matchi.ng that of cables
already in the duct, wedging is less likely to occur than
with a standard, smaller diameter optical fibre cable.
Once installed, optical fi.bre members such as 21
and 31 shown in Figure 2 or 3 are inserted into as many of
the ductlets 12 as is requi.red.
Instead of the aforedescribed fibre members 21
and 31 of near circular cross-section, the fibre members
l;~S~39~7
may, for example, be so-called ribbons, in which a thin,
wide sheath encloses an optical fibre or a plurality of
optical fibres lying in the same plane.
Manufacture of the conduit 11 is cheap compared
to the optical fibres in the fibre members 21 or 31 which
it is designed to carry, and spare ductlets 12 for future
expansion can readily be incorporated at -the extrusion
stage of the conduit 11 without adding unduly to the
overall cost. The conduit may be manufactured by adapting
conventional cable manufacturing processes such as, for
example, extrusion.
A gas flowing past the surface of a solid object
produces a drag force which largely depends on the
velocity of the gas relative to the surface. The
applicants have found that this drag force can be made
sufficiently large to pull a lightweight optical fibre
member 21, or 31 into a tubular pathway such as, for
example, a ductlet 12 of the aforementioned conduit 11.
In experiments, the flow velocity, or the flow
rate, of air through a given pathway has been found to
depend approximately linearly on the pressure difference
between opposite ends of the pathway, with the slope of
the dependency indicating that flow at useful flow rates
is predominantly turbulent.
For a given pres5ure difference, the flow rate
varies with the size of the free cross-sectional area of
the bore, while the drag force on a fibre member present
in a bore varies with the flow rate and the surface area
of the fibre member. The drag force has been optimized in
experiments by varying these parameters and, in
particular, by choosing an appropriate ratio of bore
diameter to fibre member diameter.
Experiments have been performed using a bore
diameter of 7mm. The optimum fibre member diameter for
this bore size has been found to lie between 2.5 and 4mm.
A pressure below 80 p.s.i. (approximately 5.6 kgs/cm2),
usually about 40 p.s.i. has been found sufficient to
insert fibre members of up to 3.5 gram per metre (gr/m)
lZS8390
over lengths of 200 metres. A fibre member with 2 gr/m is
easily installed over this length.
The theoretical value for the drag forces for
these dimensions has been calculated in the manner
described below with reference to Figure 6 to be 2.5 gr/m.
Lower practical values are believed to be due to the
tendency of the fibre members 21, 31 to acquire "set"
while on the supply reel. This set wou,ld appear to force
the fibre member 21, 31 against the wall of the bore 12,
thereby increasing friction.
Suitable texturing or shaping of the fibre
member surface may lead to drag forces higher than those
presently experienced.
It should be noted here that using fluid drag to
insert fibre members into tubular pathways differs
significantly from the method described in the above-
mentioned article of insetting pull cords by means of
parachutes. The parachute is propelled by the pressure
difference between the air in front of and the air behind
the parachute, and the velocity of the air relative to the
advancing cord is only minimal and the pulling force is
localized at the point of attachment of the parachute. In
contrast, using fluid drag requires a relatively high flow
o fluid relative to the surface of the fibre members.
Also, unlike the use of parachutes or potential
other methods of insetting fibre members into the tubular
pathways, using fluid drag produces a uniformly
distributed pulli~g force on the fibre member. This
reduces the strain on the optical fibres within the fibre
member to very low values.
In ordinarily pulling a fibre member through a
bend enclosing an angle 0, the tension of the leading end,
T2 is related to the tension Tl at the trailing end
T2/Tl = e ~ where ~ is the coefficient of friction. Even
a small number of bends in the pathway may therefore
result in an unacceptably high force being required at the
leading end if locking of the fibre member is to be
avoided. In contrast, the distributed pulling force
iZ5~3390
produced by fluid drag is applied evenly along the fibre
member, including in bends, and permits a large number of
bends to be easily and speedily negotiated without any
undue stress on the fibre member.
Figure 4 illustrates apparatus for ~eeding fibre
members into tubular pathways such as the ductlets 12 of
the conduit 11 of Figure 1. The apparatus consists of a
feedhead 41 which contains a straight bore 44 connected at
one end, its outlet end 42, to a flexible tube 49, and at
the other end, its inlet end 43, to a supply reel tnot
shown). The head 41 also contains an inlet 45 for air.
The outlet end 42 and the bore 44 are substantially larger
in cross-sectional area than fibre member 46. The aperture
of the inlet end 43 is only slightly larger in cross-
sectional area than that of the fibre member 46. This
arrangement forms an air block which presents a relatively
large flow resistance to air and helps prevent air
escaping through the inlet duct 43. The tube 49 is
inserted into one of the ductlets of the conduit 11.
Suitable seals between the feedhead 41 and the tube 49,
and the tube 49 and the ductlet 12 prevent undesirable
escape of the air.
In use the fibre member 46 is fed into the inlet
end 43 of the feedhead 41 by means of a pair of rubber
drive wheels 47 and 48, driven by a constant torque
driving mechanism tnot shown). Air is fed into the bore
44 through the air inlet 45 and hence is directed through
the tube 49 into the ductlet 12. The optical fibre member
46 is pushed through the inlet end 43 of the feedhead into
the bore 44 and onwards into the tube 49. Pushing of the
fibre member 46 continues until the surface area of the
fibre member which is exposed to the air flow is
sufficiently large to produce a drag force to cause the
further advance of the fibre member 46 through the tube 49
and the ductlet 12, while the rate of feed is controlled
by means of the aforementioned rubber drive wheels 47 and
48.
~2S~3~0
11 -
Figure 5 shows a branching connection between an
optical fibre trunkline 51 and a branch line 52, each
comprising a conduit 53 and 54 respectively and one or
more fibre members 55 and 56. Since, as described above,
5 the fibre members are individually inserted into the
ductlets of the trunkline conduit 53, individual fibre
members 55 can be routed from the trunk conduit 53 into
the branch conduit 54 as required, ,while other fibre
members 56 continue to the adjacent section 53a of the
10 trunkline conduit.
Referring now also to Figure 6, the drag force
on the fibre member 64 within the bore 63 of a ductlet, or
tube, 62 on account of turbulent air flow through the bore
63 can be calculated as discussed below.
These calculations show that what has been
called fluid drag or drag force above is, in fact, a
composite force, of which the major proportion is normally
due to viscous drag, and at least one other important
component due to a hydrostatic force, f' below. It will
be appreciated that the exact composition of the drag
force does not affect the principles of the invention but
the more detailed analysis below can be used to optimize
the parameters involved in carrying out the invention, and
to obtain some guidance for trial and error experiments.
The pressure difference between the tube ends
can be equated to a shear force distributed over the inner
surface of the bore 63 and the outer surface of the fibre
member 64. Thus, one has, for a small element of
length a 1 producing a pressure drop ~p
~ p~r (r2 - rl) = F (1)
where r2 = outer tube bore radius, rl = inner tube radius
and F is the viscous drag force on the inner and outer
walls of the elemental length.
If it is now assumed that the force F is
distributed evenly over the area of the inner and outer
walls, that is to say the external wall of the fibre
member and the internal wall of the ductlet respectively,
1~513390
12
the drag force, f, on the fibre member per unit length
~ = F ~ 1 ~=Qp1rr](r2 ~ rl) (2)
which gives, in the limi.t, the drag force on the fibre
member per unit length.
1 2 1) dl
In addition, we must consider the force produced
by the pressure difference acting on the cross-sectional
area of the fibre member. This is locally proportional to
the pressure gradient and therefore is distributed over
the installed length of the fibre member in the same way
10. as the viscous drag force, leading to an additional force
f, Qp~r2 (4)
Ql
giving a total force per unit length of
TOT dl~rlr2 (5)
In order to get an i.nitial estimate of this
force per unit length it is assumed that the pressure
drops linearly over the length of the bore, whether filled
by the fibre member or not. Equation 5 is then plotted,
for the case of the 6mm bore diameter with 2.5mm O.D.
fibre member, in Figure 8, for a length of 300m. Since
pressure is normally quoted in p.s.i. it has been retai.ned
here for the sake of convenience.
Coefficients of friction of around 0.5 have been
measured for the polyethylene and polypropylene fibre
members against a polyethylene bore wall. ~herefore, with
a fibre member weighing 3gms/m we could expect to install
a 300m length with around 55 p.s.i. pressure. Any extra
drag force over that required to overcome friction would
appear at the start end as a gradually increasing tension
in the fibre member as installation proceeds.
Fi~ure 7 shows in diagrammatic form the
arrangement of the modi.fied drive unit discussed with
reference to Figure 4, in whi.ch the only major change lies
. ~
125l~339(~
13
in incorporating the drive wheels 77 and 78 within the
feedhead 71.
As the foregoing discussion with reference to
Figure 6 has illustrated, the viscous drag force is
accompanied by a hydrostatic force, the force f' of
equation 5 above. This force f' has been found to oppose
the insertion of the fibre member into the drive unit,
making the incorporation of the drive wheels 77 and 78
into the drive unit preferable. The force f', referred
above as the hydrostatic potential must be overcome when
introducing the fibre member into the pressurized areas.
The drive wheels would be driven by a torque just
sufficient to overcome this potential.
The drive wheels are incorporated into the
pressuri~ed cavity 74 and thus the force on the fibre
member necessary to overcome the hydrostatic potential is
tensile. If the wheels were outside the drive unit, this
force would be compressive, and there would be tendency
for the fibre member to buckle.
For convenience the drive unit may be made to
split along the fibre member axis, and perpendicular to
the diagram, or in some other plane. The air seals 72, 73
may be, for example, rubber lips, or narrow channels.
In operation, a fibre member 76 fed into the
drive unit would be automatically taken up by the drive
wheels with just enough force to overcome the hydrostatic
potential, and fed on along the ductlet 12. The fluid
drag of the air flowing down the ductlet 12 causes the
fibre member 76 to be pulled along the ductlet 12 as the
installation proceeds. This means that such a drive unit
can be placed between two adjoining sections of conduit so
that a fibre member emerging from a ductlet in the first
conduit can be fed into the appropriate ductlet of the
second. Thus, an installation could consist of a fibre
member 76 running through a number of conduits using two
or more drive units in tandem, possibly without
supervision.
lZ~B39~
14
It will be appreciated that it is possible to
blow compounds in liquid or powder form along the ductlet
prior to, or during installation in order to provide
lubrication for the fibre members. Powdered talc is an
s example of a suitable lubricant.
The ductlets may, for example, also be formed in
a power cable, or in a conventional subscriber line, to
allow subsequent installation of opti,cal fibre members.
In the latter case, to avoid ingress of water, the ductlet
may be sealed until the time of installation of the fibre
members.
,
