Electromagnetics, Light waves & Type of Fibers
Transmitting
Light on a Fibre:
An optical fibre is a
very thin strand of silica glass in geometry quite like a human hair. In
reality it is a very narrow, very long glass cylinder with special
characteristics. When light enters one end of the fibre it travels (confined
within the fibre) until it leaves the fibre at the other end. Two critical
factors stand out:
1. Very little light
is lost in its journey along the fibre.
2. Fibre can bend
around corners and the light will stay within it and be guided around the
corners.
As shown in Figure,
an optical fibre consists of two parts: the core and the cladding. The core is
a narrow cylindrical strand of glass and the cladding is a tubular jacket
surrounding it. The core has a (slightly) higher refractive index than the cladding. This means that the boundary
(interface) between the core and the cladding acts as a perfect mirror. Light
travelling along the core is confined by the mirror to stay within it - even
when the fibre bends around a corner. When light is transmitted on a fibre, the
most important consideration is “what kind of light?” The electromagnetic
radiation that we call light exists at many wavelengths.11 These wavelengths go
from invisible infrared through all the colours of the visible spectrum to
invisible ultraviolet. Because of the attenuation characteristics of fibre, we
are only interested in infrared “light” for communication applications. This
light is usually invisible, since the wavelengths used are usually longer than
the visible limit of around 750 nanometers (nm).12 If a short pulse of light from a source such
as a laser or an LED is sent down a narrow fibre, it will be changed (degraded)
by its passage down the fibre. It will emerge (depending on the distance) much
weaker, lengthened in time (“smeared out”), and distorted in other ways. The
reasons for this are as follows:
Attenuation
The pulse will be
weaker because all glass absorbs light. More accurately, impurities in the
glass can absorb light but the glass itself does not absorb light at the
wavelengths of interest. In addition, variations in the uniformity of the glass
cause scattering of the light. Both the rate of light absorption and the amount
of scattering are dependent on the wavelength of the light and the
characteristics of the particular glass. Most light loss in a modern fibre is
caused by scattering.
Maximum Power
There is a practical
limit to the amount of power that can be sent on a fibre. This is about half a
watt (in standard single-mode fibre) and is due to a number of non-linear
effects that are caused by the intense electromagnetic field in the core when
high power is present.
Polarisation
Conventional
communication optical fibre is cylindrically symmetric but contains
imperfections. Light travelling down such a fibre is changed in polarisation.
(In current optical communication systems this does not matter but in future
systems it may become a critical issue.)
Dispersion
Dispersion occurs
when a pulse of light is spread out during transmission on the fibre. A short
pulse becomes longer and ultimately joins with the pulse behind, making
recovery of a reliable bit stream impossible. (In most communications systems
bits of information are sent as pulses of light. 1 = light, 0 = dark. But even
in analogue transmission systems where information is sent as a continuous
series of changes in the signal, dispersion causes distortion.) There are many
kinds of dispersion, each of which works in a different way, but the most
important three are discussed below:
1. Material
dispersion (chromatic dispersion)
Both lasers and LEDs
produce a range of optical wavelengths (a band of light) rather than a single
narrow wavelength. The fibre has different refractive index characteristics at
different wavelengths and therefore each wavelength will
travel at a different speed in the fibre. Thus, some wavelengths arrive before
others and a signal pulse disperses (or smears out).
2. Modal dispersion
When using multimode
fibre, the light is able to take many different paths or “modes” as it travels
within the fibre. This is shown in Figure 15 on page 33 under the heading
“Multimode Step-Index”. The distance traveled by light in each mode is different
from the distance travelled in other modes. When a pulse is sent, parts of that
pulse (rays or quanta) take many different modes (usually all available modes).
Therefore, some components of the pulse will arrive before others. The
difference between the arrival time of light taking the fastest mode versus the
slowest obviously gets greater as the distance gets greater.
3. Waveguide
dispersion
Waveguide dispersion
is a very complex effect and is caused by the shape and index profile of the
fibre core. However, this can be controlled by careful design and, in fact,
waveguide dispersion can be used to counteract material dispersion as will be
seen later.
Noise
One of the great
benefits of fibre optical communications is that the fibre doesn't pick up noise
from outside the system.
However,
there are various kinds of noise that can come from components within the
system itself. Mode partition noise can be a problem in single-mode fibre and
modal noise is a phenomenon in multimode
fibre. None of these effects are helpful to engineers wishing to transmit
information over long distances on a fibre. But much can be done about it.
1.
If
you make the fibre thin enough, the light will have only one possible path -
straight down the middle. Light can't disperse over multiple paths because
there is only one path. This kind of fibre is called single-mode fibre.
2.
The
wavelength of light used in a particular application should be carefully chosen,
giving consideration to the different attenuation characteristics of fibre at
different wavelengths. Cost considerations are important here too. The
wavelength at which a particular light source or detector can operate is
determined very largely by the materials from which it is made. Different
materials have very different cost structures. In general, the shorter the
wavelength the lower the cost.
3.
Types
of dispersion that depend on wavelength can of course be minimised by minimising
the spectral width of the light source. All light sources produce a range or
band of wavelengths rather than a single wavelength. This range is usually
called the “spectral width” of the light source.
Lasers are commonly
thought to transmit light at one wavelength only. But this is not exactly true.
Simple semiconductor lasers built for communications applications typically
transmit a range of wavelengths of between 1 nm and 5 nm wide. More
sophisticated communications lasers can produce an (unmodulated) spectral width
of as little as 0.01 nm. It is possible
to construct Light Emitting Diodes (LEDs) that emit light within only a narrow
range of wavelengths. Typical communications LEDs have a spectral width of
between 30 nm and 150 nm. The narrower the spectral width of the source the
smaller the problem of dispersion.
4. Material
dispersion and waveguide dispersion are both dependent on wavelength. Waveguide
dispersion can be controlled (in the design of the fibre) to act in the
opposite direction from material dispersion. This more or less happens
naturally at 1300 nm but can be adjusted to produce a dispersion
minimum in the 1500
nm band. This is a result of the core profile and refractive index contrast.
Achieving this balance at 1300 nm was one reason for the profile of standard
single-mode fibre. 1300 nm was a good wavelength because in the late 1980s
lasers that operated in the 1310 nm band were relatively easy to make compared
to longer-wavelength types. Indeed, this distinction with its concomitant cost difference
still exists today.
To make a fibre we
need the core and the cladding to have different refractive indices. So we need
to modify the RI of fused silica. One advantage of the liquid-like structure of
glass is that we can mix other materials into it in almost any proportion and
concentration. We are not limited to fixed ratios as we might be with
crystalline structures. Also a very wide range of materials will mix (it is
almost correct to say dissolve”) in the
glass. People have been mixing other materials into glass to change the RI for
manyyears! “Lead Crystal”
glass (used in chandeliers, vases and tableware) has a high proportion of lead
oxide mixed into it in order to increase the RI and make it “sparkle”.
We can construct an
optical fibre either by doping the core with something that increases the RI or
by doping the cladding with something that decreases the RI. However, it is not
quite as simple as this. These dopants also change other characteristics of the
glass such as the coefficients of thermal
expansion. If the core and cladding have significantly different
coefficients of expansion they may crack apart after a time - or indeed they
may crack apart during manufacture.
In optical fibre we
usually mix a proportion of germanium dioxide (4% to 10%) with the silica to
increase the RI when required. RI can be decreased by adding boron trioxide (B²O³).
There are many other materials available that can be used as dopants in this
way.16 There are many other
substances that either increase or decrease the RI. Each of these also has an
effect on characteristics of the material other than the RI (such as the
coefficient of expansion). Other common dopants used to increase the RI are
phosphorus pentoxide (P²Oµ), titanium dioxide
(TiO²) and aluminium oxide (Al²O³).
Scattering :
Most of the
attenuation in fibre is caused by light being scattered by minute variations
(less than 1/10th of the wavelength) in the density or composition of the
glass. This is called “Rayleigh Scattering”. Rayleigh scattering is also the
reason that the sky is blue and that sunsets are red.
In fibre, Rayleigh
scattering is inversely proportional to the fourth power of the wavelength!
This accounts for perhaps 90% of the enormous difference in attenuation of
light at 850 nm wavelength from that at 1550 nm. Unfortunately, we can't do a
lot about Rayleigh scattering by improving fibre manufacturing
techniques. There is
another form of scattering called “Mie Scattering”. Mie scattering is caused by
imperfections in the fibre of a size roughly comparable with the wavelength.
This is not a significant concern with modern fibres as recent improvements in
manufacturing techniques have all but eliminated the problem. The absorption
peak shown in Figure is centered at 1385
nm but it is “broadened” by several factors including the action of ambient
heat. This absorption is caused by the presence of the -OH atomic bond, that
is, the presence of water. The bond is resonant at the wavelength of 1385 nm.
Water is extremely hard to eliminate from the fibre during manufacturing and
the small residual peak shown in the diagram is typical of current, good
quality fibres. In the past this peak was significantly greater in height than
shown in the figure (up to 4 dB/km).
Fibre Transmission
Windows (Bands)
Characteristics of early fibre
Transmission Windows.
The upper curve shows the absorption characteristics of fibre in the 1970s. The
lower one is for modern fibre.In the early days of optical fibre communication, fibre
attenuation was best represented by the upper curve in Figure (a large difference from today). Partly for
historic reasons, there are considered to be three “windows” or bands in the transmission
spectrum of optical fibre. The wavelength band used by a system is an extremely
important defining characteristic of that system.
Short Wavelength Band
(First Window)
This is the band around 800-900 nm.
This was the first band used for optical fibre communication in the 1970s and
early 1980s. It was attractive because of a local dip in the attenuation
profile (of fibre at the time) but also (mainly) because you can use low cost
optical sources and detectors in this band.
Medium Wavelength
Band (Second Window)
This is the band around 1310 nm which
came into use in the mid 1980s. This band is attractive today because there is
zero fibre dispersion here (on single-mode fibre). While sources and detectors
for this band are more costly than for the short wave band the fibre
attenuation is only about 0.4 dB/km. This is the band in which the majority of
long distance communications systems operate today.
Long Wavelength Band
(Third Window)
The band between about 1510 nm and
1600 nm has the lowest attenuation available on current optical fibre (about
0.26 dB/km). In addition optical amplifiers are available which operate in this
band. However, it is difficult (expensive) to make optical sources and
detectors that operate here. Also, standard fibre disperses signal in this band.
In the late 1990s this band is where
almost all new communications systems operate.
Transmission Capacity
The potential transmission
capacity of optical fibre is enormous. Looking again at both the medium and long wavelength bands are
very low in loss. The medium wavelength band (second window) is about 100 nm
wide and ranges from 1250 nm to 1350 nm (loss of about .4 dB per km). The long
wavelength band (third window) is around 150 nm wide and ranges from 1450 nm to
1600 nm (loss of about .2 dB per km). The loss peaks at 1250 and 1400 nm are
due to traces of water in the glass. The useful (low loss) range is therefore
around 250 nm. Expressed in terms of analogue bandwidth, a 1 nm wide waveband
at 1500 nm has a bandwidth of about 133 GHz. A 1 nm wide waveband at 1300 nm
has a bandwidth of 177 GHz. In total, this gives a usable range of about 30
Tera Hertz (3 × 1013 Hz). Capacity depends on the modulation
technique used. In the electronic world we are used to getting a digital
bandwidth of up to 8 bits per Hz of analog bandwidth. In the optical world,
that objective is a long way off (and a trifle unnecessary). But assuming that a
modulation technique resulting in one bit per Hz of analog bandwidth is
available, then we can expect a digital bandwidth of 3 × 1013
bits per
second. Current technology limits electronic systems to a rate of about 10
Gbps, although higher speeds are being experimented with in research. Current
practical fibre systems are also limited to this speed because of the speed of
the electronics needed for transmission and reception.
Operational
Principles
·
. Multimode Step-Index
·
. Multimode Graded-Index
·
. Single-Mode (Step-Index)
The difference between them is in the
way light travels along the fibre. The top section of the figure shows the
operation of “multimode” fibre. There are two different parts to the fibre. In
the figure, there is a core of 50 microns (μm) in diameter and a
cladding of 125 μm in diameter. (Fibre size is normally
quoted as the core diameter followed by the cladding diameter. Thus the fibre
in the figure is identified as 50/125.) The cladding surrounds the core. The
cladding glass has a different (lower) refractive index than that of the core,
and the boundary forms a mirror.
This is the effect
you see when looking upward from underwater. Except for the part immediately
above, the junction of the water and the air appears silver like a mirror.
Light is transmitted (with very low loss) down the fibre by reflection from the
mirror boundary between the core and the cladding. This phenomenon is called “total internal
reflection”. Perhaps the most important characteristic is that the fibre will
bend around corners to a radius of only a few centimetres without any loss of
the light.
Figure Multimode Step-Index Fibre
The expectation of
many people is that if you shine a light down a fibre, then the light will
enter the fibre at an infinitely large number of angles and propagate by
internal reflection over an infinite number of possible paths. This is not
true. What happens is that there is only a finite number of possible paths for
the light to take. These paths are called “modes” and identify the general
characteristic of the light transmission system being used. Fibre that has a core diameter large enough
for the light used to find multiple paths is called “multimode” fibre. For a
fibre with a core diameter of 62.5 microns using light of wavelength 1300 nm,
the number of modes is around 400 depending on the difference in refractive
index between the core and the cladding. The problem with multimode operation
is that some of the paths taken by particular modes are longer than other paths.
This means that light will arrive at different times according to the path
taken. Therefore the pulse tends to disperse (spread out) as it travels through
the fibre. This effect is one cause of “intersymbol interference”. This
restricts the distance that a pulse can be usefully sent over multimode fibre.
Multimode Graded
Index Fibre
One way around the
problem of (modal) dispersion in multimode fibre is to do something to the
glass such that the refractive index of the core changes gradually from the centre
to the edge. Light travelling down the center of the fibre experiences a higher
refractive index than light that travels further out towards the cladding. Thus
light on the physically shorter paths (modes) travels more slowly than light on
physically longer paths. The light follows a curved trajectory within the fibre
as illustrated in the figure. The aim of this is to keep the speed of
propagation of light on each path the same with respect to the axis of the
fibre. Thus a pulse of light composed of many modes stays together as it
travels through the fibre.
This allows
transmission for longer distances than does regular multimode transmission.
This type of fibre is called “Graded Index” fibre. Within a GI fibre light
typically travels in around 400 modes (at a wavelength of 1300 nm) or 800 modes
(in the 800 nm band).
Note that only the
refractive index of the core is graded. There is still a cladding of lower
refractive index than the outer part of the core.
Single-Mode Fibre
Figure. Single-Mode
Fibre. Note that this figure is not to scale. The core diameter is typically
between 8 and 9 microns while the diameter of the cladding is 125 microns. If the fibre core is
very narrow compared to the wavelength of the light in use then the light
cannot travel in different modes and thus the fibre is called “single-mode” or
“monomode”. There is no longer any reflection from the core-cladding boundary
but rather the electromagnetic wave is tightly held to travel down the axis of
the fibre. It seems obvious that the longer the wavelength of light in use, the
larger the diameter of fibre we can use and still have light travel in a
single-mode. The core diameter used in a typical single-mode fibre is nine
microns. It is not quite as simple as this in practice. A significant
proportion (up to 20%) of the light in a single-mode fibre actually travels in
the cladding. For this reason the “apparent diameter” of the core (the region
in which most of the light travels) is somewhat wider than the core itself. The
region in which light travels in a single-mode fibre is often called the “mode
field” and the mode field diameter is quoted instead of the core diameter. The
mode field varies in diameter depending on the relative refractive indices of
core and cladding, Core diameter is a compromise. We can't make the core too
narrow because of losses at bends in the fibre. As the core diameter decreases Chapter compared to the
wavelength (the core gets narrower or the wavelength gets longer), the minimum
radius that we can bend the fibre without loss increases. If a bend is too
sharp, the light just comes out of the core into the outer parts of the
cladding and is lost.
You can make fibre
single-mode by:
·
Making
the refractive index difference between core and cladding small enough
·
Using
a longer wavelength
Single-mode fibre
usually has significantly lower attenuation than multimode (about half). This
has nothing to do with fibre geometry or manufacture. Single-mode fibres have a
significantly smaller difference in refractive index between core and cladding.
This means that less dopant is needed to modify the refractive index as dopant
is a major source of attenuation. It's not strictly correct to talk about
“single-mode fibre” and “multimode fibre” without qualifying it - although we
do this all the time. A fibre is single-moded or multi-moded at a particular
wavelength. If we use very long wave light (say 10.6 nm from a CO² laser) then even most MM fibre would be single-moded for
that wavelength. If we use 600 nm light on standard single-mode fibre then we
do have a greater number of modes than just one (although typically only about
3 to 5).There is a single-mode fibre characteristic called the “cutoff
wavelength”. This is typically around 1100 nm for single-mode fibre with a core
diameter of 9 microns. The cutoff wavelength is the shortest wavelength at
which the fibre remains single-moded. At wavelengths shorter than the cutoff
the fibre is multimode. When light is introduced to the end of a fibre there is
a critical angle of acceptance. Light entering at a greater angle passes into
the cladding and is lost. At a smaller angle the light travels down the fibre.
If this is considered in three dimensions, a cone is formed around the end of
the fibre within which all rays are contained. The sine of this angle is called
the “numerical aperture” and is one of the important characteristics of a given
fibre.
Single-mode fibre has
a core diameter of 4 to 10 μm (8 μm is typical). Multimode fibre can have many core
diameters but in the last few years the core diameter of 62.5 μm in the US and 50 μm outside the US has
become predominant. However, the use of 62.5 μm fibre outside the
US is gaining popularity - mainly due to the availability of equipment
(designed for the US) that uses this type of fibre.
Connectors
In many situations it
is highly desirable to be able to change configurations easily. This means that
we want to plug a cable into a wall socket and to conveniently join sections of
cable. Many different types of connectors are available to do just this. They
hold the fibre ends in exact position and butt them together under soft pressure
to obtain a good connection. But this all depends on the precision to which
connectors can be machined. Most mechanical devices are machined to a tolerance
of around one millimetre. Fibre tolerances are around one micron. This means
that connectors have to be between 100 and 1000 times more accurately machined
than most mechanical “things”. Thus, connectors are difficult to manufacture
and hard to fit. This all results in a relatively high cost. Today's connectors
are very efficient with typical “off-the-shelf” connectors measuring losses of
around .2 dB. However, most manufacturers still suggest planning for a loss of
1 dB per connector. Connectors are rarely fitted in the field. The cost of the
equipment needed to fit most current connectors is over US $100,000 and the
process is difficult to perform in the field. Thus either you purchase cables
with connectors already fitted (the best approach) or you buy connectors
already fitted to a short length of cable. In the latter case you splice the
pigtail coming from the supplied connector to your cable in order to fit the
connector.
There are some types
of connector available that are advertised as being easy to fit in the field
but these are not the common types used on standard equipment. The alternative
of purchasing a cable with connectors already fitted is a very good one.
Connector manufacturers make protective sockets into which the cable connectors
can be inserted and locked. These protective sockets have a fitting to allow
then to be pulled through a cable duct whilst protecting the fibre cable. Of course
cable is purchased in standard lengths but this isn't too much of a problem as
excess can be coiled under the floor and gives flexibility when the need arises
to move the termination point.
Some Typical Connector
Types. Most connectors are available in simplex or duplex configurations as
shown here.
There are many
different types of connector reflecting development history and country of
origin. Connectors for MM and SM fibres are generally different but the most
popular connectors come in versions for either type. In today's world the SC
connector is fast becoming the de facto standard. It is interesting that there
are SC connector plugs but not (in general) SC sockets! The connectors are
built to be inserted into a sleeve which accepts one connector at either end.
Thus two plugs are connected together with a sleeve. This significantly increases
the versatility of the system. Of course sockets are used for direct mounting
onto a piece of equipment but in general they are not used on the end of cables.
One valuable feature of the SC connector is that the plug comes with a plastic
cover over the fibre end. This cover is pushed aside as the plug is inserted into
the connecting sleeve or socket. This helps protect the cable end when it is not
connected and also it is an excellent aid to eye safety.
Many communications
standards specify which optical connector should be used with a particular
system.
The MT connector
joins multiple fibres in a flat cable. In the illustrated case there are eight
fibres. The connectors plug together and then are inserted into a holding clip
which applies pressure to hold the two connectors together.
Fibre Cables
As we have seen,
fibres themselves are generally 125 μm in external
diameter (very small indeed). While they are very strong under tension
(stronger than steel in fact) they break very easily when subject to lateral
pressure or any kind of rough handling. To make practical use of fibre for
communication the fibre needs to be enclosed in a cable.
Fibre cables vary
widely in their characteristics due to the differing environments in which they
are installed and requirements they must fulfill. Fibre cables are made to suit
the application they are to perform and there are hundreds, perhaps thousands
of types. Indeed, if you want to construct an outdoor fibre cable link for a
few hundred kilometers you can go to the cable manufacturers and specify anything
you want. All details of cable construction are negotiable. The objective of
the cable is to protect the installed fibre from anything that may damage it:
Tensile Stress
While fibre itself is
quite strong under tension, stress causes a significant increase in attenuation
and a number of other undesirable effects. We need to protect the fibre from any
kind of stress.
Bends
Bends in the fibre
that are too small in radius cause signal loss. Microbends in the fibre caused
by crimping of the cable also cause signal loss. One function of the cable is
to prevent the fibre being bent to a radius where loss may occur. With long
distance outdoor or undersea cables this is not a big problem.
Such cables often
have a minimum bend radius of a few feet!
Physical Damage from
Environmental Conditions
Just what is needed
to protect the cable varies with the particular environment. In many indoor
environments vermin (rats, etc.) chew cable (they usually find electrical cable
unpleasant but fibre is less so). In outdoor ones, gophers and termites also
like eating cable. Heavy earth-moving equipment also has very little respect
for cable integrity. One major hazard for outdoor cables is cable-laying
machines. In many countries cables are laid along defined cable
“rights-of-way”. When someone comes to lay a new cable along a route where
there are already other cables, the existing ones tend to get cut.
Damage in the Cable
Installation Process
Cable doesn't just
have to operate satisfactorily in its installed environment but it must
withstand the stresses of being installed. In some cases these stresses can be
quite severe, for example, being lifted up the core of a multi-story building
or dragged through a long conduit.
Water
It sounds illogical,
but waterproofing is often more important in the fibre optical environment than
it is in the electrical world! Glass immersed in water gradually picks up
hydroxyl ions. As mentioned earlier, the presence of hydroxyl ions greatly
increases the absorption of light. In addition, the presence of water causes
micro-cracking in the glass and this causes the scattering of light. The micro-cracks
also weaken the fibre significantly. Water is the worst enemy of an optical
fibre system.
Lightning Protection
Lightning is a
problem for all outdoor cables containing conductive materials. This, of
course, depends on which part of the world you happen to be in. In some places
lighting can hit the ground and sever an underground telecommunications cable
up to 10 metres away! In addition there are other functions that need to be
supported in some environments. For example in some cable situations (especially
undersea) it is necessary to provide power for repeaters or amplifiers along a
long-distance cable route. One example of this is in submarine cables.
Electrical power cabling is often included to deliver power to the repeaters.
Cabling Environments
There are many
different environments in which we wish to install fibre cable. Cables are
specially designed for each environment:
Outdoor Buried Cable
(Long Distance)
Typical outdoor
buried cables contain a large number of single-mode fibres (up to 100). They
contain very extensive waterproofing, strength members and often armouring.
Outdoor Buried Cable
(Campus Area)
These are typically
lighter than the long distance variety and contain both multimode and
single-mode fibres. These usually have good waterproofing and protection but it
is often not as strong as long distance varieties. In some places cable is
installed in a conduit such as a 2-inch diameter steel pipe. In this case you
don't need strong armouring.
Outdoor Overhead
Cable
Cable intended for
overhead use needs to have very great tensile strength to prevent the fibres
being stressed. Typically they have a separate support member which takes the
stress outside of the cable itself.
Outdoor Overhead
(High-Voltage Earthwire) Cable
One very popular and
creative place to put optical fibre is inside the earth wire of a high voltage
electrical transmission system. A common system of this kind might operate at
132,000 volts. The earth wire is usually the top wire on the tower (relatively
safe from vandalism). You always have to have a name and an acronym. These
cables are called “Optical Ground Wire” (OPGW) cables.
Undersea Cable
The undersea
environment is the most difficult cabling environment imaginable. Keeping high
pressure salt water out of the cable poses a very significant challenge. In
contrast to the large numbers of fibres in terrestrial long distance cables
there are usually only a small number (6-20) of fibres in an undersea cable.
Undersea cables also often carry electric wires to provide power to
regenerators (repeaters or amplifiers) in the cable.
Indoor Cabling
Typically indoor
cables have a very small number of fibres (most often only two) and these are
generally multimode. In the indoor office environment there is less need for waterproofing
or armouring than in the outdoor environment. However, you do need some
protection from vermin (such as rats) and from accidental damage both during
and after installation. In addition, it is often desirable to make the cable
from materials that don't give off toxic fumes when they burn. This costs more
but may save lives in the case of a fire. Indoor cables are typically short
distance (up to 300 metres or so) and need to be relatively light and flexible
for installation in the office environment. They are usually terminated with
pluggable connectors.
Jumper Leads and Fly
Cables
One of the harshest
environments for a fibre cable is as an end-system connector in the office.
These often tend to run across the surface of a floor. They get stepped on and
desks and chairs occasionally run their wheels over them. These are the best
ways known to science of breaking a fibre.
The First Stage -
Coating the Fibre
The basic form of
optical fibre is called “primary coated optical fibre” or PCOF. This is the
fibre as it emerges from the drawing tower. You have the core and cladding of
course surrounded by a protective plastic jacket. In this form the overall diameter
is 250 microns. In many applications (such as in “loose-tube” cables) there is
a need to identify the fibre within its cable. To allow this, either the jacket
itself is coloured or there is a further very thin coloured coating added
immediately prior to building the cable. Since there are no standards for
colour coding, cable makers tend to
prefer to colour the fibre at cabling time to allow for flexibility in satisfying
customer specifications.
In end-user
environments PCOF is never used without further encasement in a cable. However,
in many optical research and development laboratories it is common to see PCOF
running around the laboratory attached to walls and ceilings with sticky tape.
This is not recommended in normal office environments!
Secondary Coated
Optical Fibre (SCOF)
For cabling within
office environments, PCOF is usually further coated with a secondary coating as
shown in Figure. This is called “Secondary Coated Optical Fibre” (SCOF). The
secondary coating forms a tight bond with the primary coating so that when you
want to strip the fibre to make a join you have to deal
with a coating that
is 900 microns in diameter. Other levels of cladding within the cable will peel
away from the SCOF relatively easily.
Basic Cable
Construction
Loose Tube
Construction
Cables can be
classified into three types depending on how the fibre is encased within them:
Tight Buffered
Construction
Tight buffering is
where the secondary coated fibre (SCOF) is encased firmly in surrounding
material as shown in Figure. This is similar to electrical cabling.
Construction of this type is usually used for indoor applications where the
number of fibres needed in the cable is low and distances are relatively short.
However, it is also used for medium distance outdoor applications such as
around a campus.
Loose Tube
Construction
In loose tube
construction a small number of PCOF fibres are carried inside a plastic (PVC)
tube of 4-6 mm in diameter. Typically between one and eight primary coated
fibres are carried in a single tube. There is plenty of room in the tube for
the fibres to move loosely within it.
The idea is that you
use somewhat more fibre than cable (5%-10%) in each tube. Fibre coils around
inside the tube in a helical pattern. If the cable is stretched or bent then
the fibres inside do not experience tension.
Loose Tube with Gel Filler
In most buried
outdoor cable today we use a loose tube construction where the tube is filled
with a jelly. This prevents the ingress of water from faults in the cable. It
also buffers the fibres from one another and helps to prevent losses due to microbends
caused by irregularities on the surface of the insides of the tubes. The
composition of the gel used is a significant design issue. Historically, petroleum
jelly was used but this exhibits significant changes in viscosity with
temperature. Viscosity is very important for a number of reasons:
·
The
fibre must be free to move within the tube to counter stress caused by
temperature changes and/or cable laying.
·
The
viscosity needs to be high enough to provide some mechanical stability to the
tube. And it should allow the cable to be run vertically without the gel
settling down to the bottom.
·
It
needs to be reasonably well-behaved in the field when a cable has been severed
and needs to be re-joined. For example, if the gel became very mobile on a hot
day and the cable was severed, the gel could run out of the tubes for a long
distance. . Of course the gel needs to be stable
during the process of cable manufacture where it may be subject to high
temperatures.
Silicone gels are a
lot better generally than petroleum ones but today specially designed synthetic
gels are used which all but eliminate the early problems with changes in
viscosity.
Indoor Cables
Single-Core Cable
The most basic form
of indoor optical cable is shown in Figure . This is simply a single strand of
SCOF with a layer of strengthening aramid (or fibreglass) fibres and an outer
PVC jacket added. Single-core cable of this nature is used in short lengths as
jumper or fly cables but is almost never used for fixed cabling. This is because
it is lower in cost to have a cable with many fibres in it than it is to have many
single-core cables. In any case you almost always need two cores anyway.
Dual Indoor Cable
The construction
shown in Figure is a very common low
cost indoor cable construction. Two basic single-core cables are carried
together in a common
6-Core Tight Buffered
Indoor Cable
Figure shows the cross-section of a typical
heavy-duty, tight buffered indoor cable. Such a cable might be installed
vertically in a building riser connecting many floors. Many simple single-cored
cables are encased in a common sheath. The central strength member (in this
case plastic) supports the weight of the total cable. Tight buffering ensures
that individual fibres are not put under tension due to their own weight. Cables
of this kind commonly come with up to 12 fibres each of which is individually
sheathed and coloured.
Air Blown Fibre (ABF)
In a “blown” fibre
installation instead of installing fibre cables you install narrow plastic
tubes or conduits. Later, very lightly clad fibre bundles are installed into
the tubes (blown in) using a system of compressed air (or compressed nitrogen).
There are various blown fibre systems available but one of the popular ones
allows bundles of between 2 and 18 fibre strands (either SM or MM) to be
inserted into the already installed tubes. This can be achieved for distances
of up to about two kilometers (or about 6000 feet). The fibre bundles blown
into the tube conduits are really very lightweight cables specifically designed
for their aerodynamics. In the installation process the fibre floats within the
tube and there is very little contact between the bundle and the tube. The
installation process is quite fast with drawing speed approaching 50 metres per
minute.
The idea of blown
fibre originated in the 1980's as a potential solution to the fact that fibre
specifications were rapidly changing and people didn't want the expense of
installing fibre cable knowing that they would want to replace it in the near
future. In today's world, fibre specifications have stabilised significantly
and there is less emphasis on the potential need for change later. However,
there are some very important advantages in the blown fibre technique.
1. When you install
the tubes you can install them in sections as convenient and join them together
later. Typical tubes can be joined using very simple “push-fit” connectors
within specially designed junction boxes. In some office environments it is
very difficult to install a long, unbroken fibre cable. In this situation
installation of the conduit tube in sections can save substantial cost. When
the fibre is “blown in” each single strand is unbroken from end-to-end. Thus
you don't have the problem (and cost) of joining the fibre. In addition, because
the fibre is installed in long, unbroken lengths you don't get losses or reflection
problems from the joins. This point is particularly important in the case where
some parts of the fibre connection are indoor and other parts outdoor. Once the
conduits are connected between the indoor and outdoor sections, the fibre is
blown in as a single unbroken cable eliminating the need for making fibre joins
(or having connectors) between indoor and outdoor sections.
2. You can install tubes
with multiple cavities so that additional fibres can be installed later as the
occasion demands. This saves some fibre cost but is more significant in the
additional flexibility provided.
3. It is possible to
remove fibre from the tubes and re-install it into other tubes on other routes
as demands change. Since installing the tube (or regular fibre cable) is the
major part of the installation cost, this allows for very low cost changes as
the installation evolves.
4. The original
reason for blown fibre while perhaps de-emphasised these days has not gone
away. As mentioned in other parts of this book, there is a strong difference of
opinion among fibre optical engineers on the future of multimode fibre. As
speeds increase the bandwidth limitations of GI MM fibre become
more and more
restrictive. Single-mode fibre itself (just the fibre) is intrinsically lower
in cost than multimode fibre but devices that connect to it are very much more
expensive. It seems generally agreed in the industry that sometime we will have
to change from MM fibre in the office to SM and when that happens those with
blown fibre installations will be able to effect the change for very
significantly lower cost than those with more conventional fibre cabling.
Outdoor Cables
It consists of six gel-filled loose tubes
supported by other cable elements designed to provide strength, mechanical
protection and protection against the ingress of water. Note that in addition
to the gel filling in the tubes carrying the fibres there is gel surrounding
these tubes within the cable. The illustrated cable has six fibres per tube for
a total fibre count of 36. This general cable geometry is used with up to
twelve tubes supporting as many as eight fibres each for a total of 96 fibres
in the cable. If an electrical power supply is needed, copper wires can replace
the fibres in one or more of the tubes.
Typical Outdoor Fiber Cable (Loose Tube - Gel
Filled)
The central strength
member is often made from a hard plastic material rather than steel. Indeed the
whole cable is often constructed from non-metallic materials. The outer layer
of armouring wire is a customer option which is only added to cable destined
for use in places where there is a significant danger of damage from the environment.
The use of stainless steel is unusual (but optional) as this is a
high-cost material.
Ordinary steel is much lower in cost but will rust if the outer sheath is
penetrated and water gets in. In some locations (such as tropical areas) it is
customary to add an outer nylon covering to prevent attack by termites.
Segmented-Core Cable
Design
An alternative loose
tube construction is shown in Figure. Here the cable is formed from an extruded
plastic member in the shape of a gear-wheel. Individual primary coated fibres
are carried in the indentations (channels) around the outside. In the
illustrated case there are six channels for fibres and up to eight fibres may be
carried in each. In some geometries up to twenty or so channels are used with
as few as one fibre
per channel. Outer elements of the cable are the same as for the regular
loose-tube
construction. This is
a simpler and thus lower cost way of making the cable.
Outdoor Aerial Cable
Aerial cables are
designed to be supported from towers and there is a significant problem with
stress. One alternative is to have a very strong central strengthening wire.
However, if you do this there is often a lot of crushing pressure on the cable at
the points of support. The design shown in Figure shows a typical cable designed for overhead
installation. It is not very different from underground cable except it is
enclosed in a common sheath with a strong separate support wire. The weight of
the fibre cable is supported evenly all the way along its length and stresses
are minimized.
12-Core Optical Flat Cable
The basic
construction shown in Figure was used extensively in the US in the early 1980's
for medium distance communication (up to 10 km or so) with multimode fibre. A
flat cable containing 12 fibres is constructed by sandwiching the fibres
between two layers of mylar tape with a glue (to hold the fibres in place) in-between.
The flat cable is only about 5 mm in width. A stack of 12 of these flat cables
forms a square cross-section which is then embedded within a strong protective
cable structure. The key to this construction was the use of a metallic
connector which terminated all 12 fibres of one flat cable in one operation.
While this worked reasonably well for multimode fibre it didn't have sufficient
precision for single mode fibres and the system is no longer used.
Undersea Cables
Typical Undersea Cable
Design
Undersea cables are
significantly different from other types of cable.They usually have a very
small number of fibres (between perhaps 4 and 20). This contrasts with
terrestrial cables which often have up to 100 fibres in them. . At great depths water is under very high pressure and
such cables have to prevent water ingress. For this reason all spaces in the
cable are filled with very dense plastic or polymer material except for the
cavity immediately around the fibres themselves. This is a gel-filled tube as
in terrestrial cable construction. The copper sheath shown in the figure is an
unbroken tube intended to help keep water out.
In contrast to what
we might expect, under the sea is not a completely safe place. Ship's anchors
and fishing trawlers can do significant damage to undersea optical cables. Thus
the undersea operators commonly make a distinction between “shallow” and “deep”
water. Water is considered “deep” if the bottom is more than 1000 metres from
the surface. Deep water is a relatively safe place to situate a cable and in
this environment it is typically laid on the sea floor without special
armouring. Thus cable operators plan the cable route to maximise the amount of
cable laid in deep water. In shallow water it is common to dig a trench in the
sea floor and bury the cable. Also, cables laid in shallow water are typically
heavily armoured.
No comments:
Post a Comment