Optical Fibre Cable

Optical Fibre Cable Page

Optical Fibre Cable

Description of the OALC-4 Cable

The image shown is what is known as Lightweight (LW) cable. LW cable is deployed in water depths between 2000 and 8000m.

In the centre of the cable is a 3mm diameter steel tube which, for transoceanic systems, can contain up to 16 fibres. Optical transmission is unidirectional, so two fibres (a fibre pair) are required for two-way transmission. Each fibre can carry many different wavelengths of light that are at the top end of the infra-red spectrum, and each fibre is currently capable of supporting a transmission rate of eighteen terabits per second (18Tbps), with the potential that 24Tbps per fibre pair will be available in the near future.

The fibres are made of extremely pure glass, with a clear primary cladding for physical protection and a secondary colour coating for identification. An example of the colour coding used is Aqua, Black, Blue, Brown, Clear, Green, Grey, Lime, Orange, Magenta Pink, Purple, Red, Tan, White and Yellow.

The steel tube is filled with a thixotropic gel to limit water ingress if the cable is damaged underwater, and to dampen fibre movement in the tube when the cable extends under tension. Around the steel tube are two layers of high tensile steel wires, laid up helically. The first layer has 8 x 1.4mm diameter wires, laid up with a left-hand lay, and the second layer has a total of 16 wires (8 x 1.0mm and 8 x 1.3mm), also laid up with a left-hand lay. These 24 wires have two functions: They provide the tensile strength of the cable, the Ultimate Tensile Strength (UTS) of this LW cable being 70kN or 7 tonnes, and they also form a vault around the fibre package, which protects it from the hydrostatic pressure at the bottom of the deep oceans.

The steel wires are laid up with a left-hand lay to aid coiling clockwise in a cable tank and are then overlaid with a longitudinal ‘C’ tube of copper, welded along the seam. This copper tube again has two functions: It provides an electrical path (1.0 or 1.7Ω per km), to provide power to the repeaters or allow electrical tests for fault location, and also provides a barrier to limit hydrogen ingress into the fibre package. Over time, hydrogen ingress will increase the optical attenuation of the fibre. On top of the copper tube a Medium Density Polyethylene (MPDE) sheath is extruded. This provides electrical insulation and limited physical protection. The result is a LW cable with an outside diameter of 17mm.

In shallower water, additional protection is added to the LW cable to protect it from external aggression. This additional protection adds to the cable’s mass and volume but increases the tensile performance and crush/impact resistance.

The History of Fibre Optic System Submarine Manufacture on the Greenwich Peninsula.

In 1953, a few miles downstream from Enderby Wharf, Charles Kuen Kao, a young student from Hong Kong, born in Shanghai in 1933, enrolled at Woolwich Polytechnic to study for his A-level exams. He stayed on at Woolwich to take his bachelor’s degree in electrical engineering, graduating in 1957. Some years after Kao left, Woolwich Polytechnic became Thames Polytechnic, which in 1992 became the University of Greenwich. Today, the university’s headquarters are at the Old Royal Naval College, upriver from Enderby Wharf, although the engineering department is at Chatham in Kent.

After graduation from Woolwich Polytechnic, Kao crossed the river to work in the microwave division of Standard Telephones & Cables (STC) at its North Woolwich factory. A few years later, Loughborough Polytechnic — today, Loughborough University — in Leicestershire offered him a lectureship. However, STC persuaded him to stay within the group by offering him a research position at its R&D centre, Standard Telecommunication Laboratories (STL) in Harlow, Essex.

The American engineer and physicist, Theodore Harold Maiman (1927-2007), was the first to demonstrate a working (ruby) laser, in 1960. By 1962, the first coherent light emission from a semiconductor laser had been produced by research teams at both General Electric and IBM, in the USA.

At STL, Kao and his colleague George Alfred Hockham (1938-2013), developed the idea that information could be carried not as radio waves or by electric currents, but in beams of laser light carried down thin fibres of glass.

A fibre of glassy material constructed in a cladded structure with a core diameter of about λ° and an overall diameter of about 100 λ° represents a practical optical waveguide with important potential as a new form of communication medium … compared with existing co-axial cable and radio systems, this form of waveguide has a large information capacity and possible advantages in basic material cost.’

Hockman and Kao

George Hockham and Charles Kao

(Courtesy of ASN)

They published their proposal in 1966, in a research paper that started the optical fibre communication revolution. The use of the term, ‘large information capacity’ proved to be a massive understatement. What Hockham and Kao had established was that the attenuation of glass fibre was not a fundamental property of the material but was caused by impurities. If a sufficient number of these impurities could be removed, then attenuations could be reduced to a few decibels per kilometre or even less. However, this was easier said than done, and it was not until the late 1970s that experimental and then commercial terrestrial fibre optic systems went into operation. Development work continued and by 1980, the first sea trial of a fibre optic submarine system containing an optical repeater was conducted by STC in Loch Fyne, Scotland.

Sea Trial Loch Fyne

Sea Trial in Loch Fyne

(Courtesy of ASN)

Twenty years after Kao and Hockham’s pioneering paper, STC supplied the first international, repeatered subsea optical fibre system, UK-Belgium No. 5. This system was 113 kilometres long and contained six optical fibres that provided three separate transmission circuits. The system contained three optical repeaters manufactured at Greenwich, and the cable was manufactured in Southampton.

Unlike in previous eras, the manufacturers had set out from the start to develop systems that could cross the deepest oceans, so the submarine cable and repeater designs were already in place for the next step, a system across the Atlantic Ocean. The first transatlantic fibre optic system was TAT-8, opened for service in 1988, and its design and installation involved a number of new technological challenges. Because only two fibres are required for two-way transmission and a cable can contain more than two fibres, it is possible to divert fibre pairs into different cables, so that a single main cable can service multiple destinations. This required a new submerged housing called a Branching Unit (BU). The BU can separate the fibre paths and also contains switching circuitry to manage the configuration of the system power feed. Later designs of BU now allow individual wavelengths to be extracted from or inserted into the main path from a spur cable. This new BU add/drop technology was first developed by Alcatel ASN for subsea scientific arrays and is now being deployed for networks to connect offshore oil and gas platforms. It is also supplied for most major telecommunications submarine cable networks, by ASN and the other major manufacturers.

Branching Unit

ASN Branching Unit Being Deployed

(Courtesy of ASN)

Having been the leading supplier during the Telephone Era of the 1950s through 1980s, STC went on to be one of the major suppliers of optical subsea systems in the world. This included the UK leg of the first transatlantic fibre optic system, TAT-8, in 1987, and all of the first transatlantic and transpacific private optical systems, PTAT-1 and NPC, in 1988 and 1990 respectively.

The History of STC

STC started life in the UK as Western Electric Ltd in 1883, a wholly-owned subsidiary of Western Electric USA, through the purchase of part of the declining submarine telegraph cable factory of W T Henley’s Telegraph Works Company Ltd in North Woolwich. Standard Telephones & Cables was founded in 1925, when the infant International Telegraph & Telephone Corporation (ITT) purchased Western Electric Ltd.

Between 1979 and 1982, as part of its restructuring, ITT sold off all but a minority shareholding in STC, and in 1982 STC was launched onto the London stock market as STC plc and STC Submarine Systems, with its headquarters at Enderby Wharf, was formed as a subsidiary. In 1987 ITT divested most of its remaining telecommunications businesses by forming a joint venture company, named Alcatel, with France’s Compagnie Générale d’Electricité. In the late 1980s, the chairman of STC plc, Sir Kenneth Corfield (1924-2016) made the term ‘Convergence’ a buzz word. A visionary ahead of his time, he masterminded a takeover of UK computer company International Computers Limited (ICL). His rationale for this move was that technologies were bound to converge and that in the future individuals would have one device to provide computer, telephone and television services. How stupid was that concept? Instead of lauding Corfield, the markets considered his strategy ridiculous and castigated him. The share price plummeted, the company went into decline, Corfield was forced out, and in 1991 STC was finally taken over by Canadian company Northern Telecom, usually abbreviated to Nortel.

Although STC Submarine Systems continued to be successful, the English and French businesses were brought together again in 1994, when Alcatel Alsthom bought STC Submarine Systems from Nortel to form ASN, with its headquarters in Villarceaux, near Paris in France.

ASN closed the STC Southampton cable factory in 1996, in favour of its Calais cable factory. This brought to an end 146 years of submarine cable manufacture in the UK. However, the Greenwich site was retained for the design, development and manufacture of repeaters, power feed equipment and other subsea network equipment, together with project management and marine services.

Optical System Capacity

The optical era can be subdivided into two transmission technology generations.

First generation optical repeaters operated through a process of detecting the incoming signal and then regenerating it as a new laser light pulse. They initially worked at a transmission wavelength of 1310 nanometres and a digital line rate of 280 megabits per second. By 1990, technology had advanced and the transmission wavelength had moved to 1550 nanometres with a digital line rate of 565 megabits per second, providing 80,000 separate voice channels, each operating at 64 kilobits per second over one fibre pair. This number of voice channels was a factor of ten better than the best that had been achieved during the Telephone Era.

The ability to transmit speech as a digital as opposed to analogue signal is yet another British invention. The technique of sampling voice frequencies and converting them into a digital code was conceived and patented in 1938 by Alec Harley Reeves (1902-71). However, the required circuitry was complex and was not commercially viable until after the invention of the transistor in 1947.

Reeves joined Western Electric Ltd in 1923, thus becoming an STC employee in 1925. After the Second World War, Reeves went to work for STL, initially at Enfield in North London and then at Harlow, where he managed the team led by Hockam and Kao.

This transmission capability of submarine cables was now in excess of the capacity available via satellite and, once again, by the end of the 1980s submarine cables became the dominant international telecommunications medium, a position that they still hold. This must have been beyond Hockham and Kao’s wildest dreams, but more was to come.

The development of the second generation of optical repeaters can be traced to 1986, when the erbium-doped fibre amplifier (EDFA) was first demonstrated by Professor David Payne and his team at Southampton University in the UK. In simple terms, the EDFA consists of a length of optical fibre doped with the rare earth erbium, which, when excited by a pump laser, amplifies the incoming transmission signal. The EDFA is much simpler and more reliable than regenerative circuitry and offers direct amplification independent of the signal line rate. It also allows for greater spacing between repeaters, reducing system costs.

From its initial invention, it took several years for Payne’s group, and a parallel development team at Bell Laboratories in the USA, to produce an EDFA that could be manufactured in volume with sufficient reliability for subsea repeaters.

Optical Cable Set

ASN OALC-4 Optical Cable Set

(Courtesy of ASN)

The first transatlantic optically amplified systems were TAT-12 and TAT-13, creating a ring network; these systems used a transmission wavelength of 1550 nanometres with a line rate of 5 gigabits per second (5 billion bits of data per second) on two fibre pairs. They went into operation in 1996. Around this time, the amount of data transmitted on subsea systems began to exceed voice traffic and the convention of expressing system capacity in 64 kilobit voice channels was abandoned.

During early experiments it was found that the EDFA could simultaneously amplify signals at two or more wavelengths, a technique known as Wave Division Multiplexing (WDM), and something that was not possible with the first generation, regenerative systems. WDM was quickly developed to offer 16 wavelengths per fibre pair. The ability to reduce the spacing between wavelengths was then developed for terrestrial systems, giving birth to Dense Wave Division Multiplexing (DWDM), which was quickly taken up by the submarine cable industry. This gave suppliers the opportunity to develop and offer systems with even more capacity, but still using only a single fibre pair.

Because of the EDFA, the concept of the ‘transparent pipe’ became popular — the idea that the capacity of a fibre system is limited only by the equipment connected to each end. This is, of course, an over-simplification, as system design is always contingent on current knowledge and the available technology. All submarine systems were, and still are, designed to have a specific ‘design capacity’ which is based on the technology available at the time. Generally, they are equipped at a lower capacity, allowing for growth over their theoretical 25-year design life. However, in a relatively short timescale, by the year 2000, the available capacity on a fibre pair for an optically amplified system had moved from one wavelength (λ) at 5 gigabits per second in the mid-1990s to an industry standard offering of 64 λ, each carrying 10 gigabits per second — providing 640 gigabits per second.

The total capacity of a submarine cable is a function of line rate, wavelengths, and the number of fibre pairs in the cable. For repeatered systems, the number of fibre pairs is constrained by the number of amplifiers that can be accommodated in the repeater housing and that can be powered through the cable. From its inception, the repeatered system model had been built around a maximum of four fibre pairs per system, but during the ‘dotcom’ boom, design and development was undertaken for six and eight fibre pair repeaters. However, even greater increases in line rates, plus advances in coherent and DWDM technology, have rendered these bigger repeaters unnecessary for most modern systems.

Optical Repeater

ASN Optical Repeater Being Deployed

(Courtesy of ASN)

Today, submarine systems capable of carrying 18Tbps per fibre pair are being deployed. The next technology steps of 150Gbps, 200Gbps, and 400Gbps channels are being marketed by system suppliers. In addition, wavelengths (λs) in the 1600nm band (L Band) are also being supplied. This increases the total λs, that can be carried on a fibre pair from more than 100 to 240, i.e., a potential capacity of 24Tbps or greater per fibre pair.

Reliability and Cable Burial

Submarine cable systems are designed to work effectively for at least twenty-five years. Extensive research, development and qualification programmes ensure that component failures in the submerged plant are almost unheard of. Therefore, the major threat to these systems comes from external aggression, whether this is natural or man-made. The risks of cable damage from natural phenomena, such as mobile sea bed sediments, bottom currents, seismic activity, turbidity currents, ice scouring and the like, can be avoided by judicious route selection. The risks of cable damage from anchors and trawls are mitigated again by route selection, as well as cable armouring and. most importantly, cable burial.

Since the beginning of the optical era, the burial of submarine systems on continental shelves to protect against external aggression has become ubiquitous. However, in an industry that is over 160 years old, simultaneous lay and plough burial of cable is a relatively young technology. For the majority of the telegraph and telephony eras, cables were surface laid and external aggression faults were managed by heavy armour cable protection and network diversity.

Although a number of attempts to bury subsea telegraph cables on the UK continental shelf were made in the early part of the 20th century, it is general accepted that the Western Union Telegraph Company was the first to develop a viable ship-towed cable plough. By the end of the 1930s, Western Union had completed development of its design and concluded that a trench depth of no more than 10 inches (25cm) would be practicable, given the tow forces that would be necessary. In 1938, this plough was deployed from the British cableship Lord Kelvin off the coast of Ireland to bury sections of three of Western Union’s transatlantic telegraph cables.

Another 20 years were to pass before, due to the number of faults caused by fishing to TAT-1 (1956) and TAT-2 (1959) on the eastern continental shelf of the USA, it became apparent to the system owners that some improved form of protection was necessary. From the early 1960s, Bell Labs, on behalf of the American Telephone & Telegraph Corporation (AT&T), developed a series of plough systems (Sea Plow I to V), which were used in the 60s and 70s to bury AT&T’s transoceanic telephone cables. These plough designs owed much to the work of Western Union; they produced an open trench depth of 24 inches (60cm) and could operate to a water depth of 500m.

In the Far East, the first ploughing of a long-haul submarine cable took place in 1976, with the installation of the East China Sea Cable (ECSC). The equipment used was a multi-blade plough, developed by the Japanese Kokusai Denshin Denwa (KDD) group and towed by the cableship KDD Maru. It could achieve a burial depth of 60-70 cm in water depths down to 200 metres.

By the early 1980s, fibre optic technology was promising subsea cable system owners an unheard-of increase in traffic-carrying capacity. At the same time, commercial fishing was becoming more intensive; trawlers were getting larger and were operating in greater water depths. This combination made system security an increasingly significant consideration. In the UK, British Telecom International (now BT), privatised from the General Post Office (GPO) by the Thatcher government in 1981, conducted a thorough investigation into the risks to submarine cables from external aggression in the English Channel, North Sea and on the Atlantic Continental Shelf. The study concluded that:

  1. It was uneconomic to bury cable to protect it against anchor faults.

  2. Subject to soil strength, a burial depth of 600mm was sufficient to give good protection against all known fishing techniques.

  3. Burial should be carried out down to the 1,000m contour.

Based on this study, British Telecom (BT), in collaboration with Soil Machine Dynamics (SMD), developed a new design of ship-towed plough. This plough was successfully sea trialled in early 1986 and was first used to install the historic UK-Belgium No. 5 system that same year. The plough design was a major step forward from the AT&T Sea Plow designs, solving a number of technical drawbacks such as cable residual tension, catenary management, ability to steer, and self-loading/unloading of the cable without cutting it. In the same year, KDDI modified its existing PLOW system in order to increase its burial depth capability to 1.5 metres.

From 1987, BT Marine and Cable & Wireless Marine were equipped with SMD ploughs and in 1992, KDDI introduced to its fleet the SMD designed and manufactured PLOW-I. All these companies were then capable of 1 metre burial depth in water depths up to 1,000 metres. By 2000, this British plough design had become the de facto industry standard and remains so today.

Over the last 18 years, through SMD in the UK and companies like Perry Tritech Inc. in the USA, development of plough technology has advanced even further. Today we have 1, 1.5, 2 and 3 metre burial depth ploughs, plus jet-assisted and rock-ripping ploughs. Ploughing on the edge of continental shelves regularly takes place in water depths of 1,500m and, where seabed conditions allow, ploughing in even greater water depths has been achieved.

Heavy Duty Plough

Heavy Duty Plough Being Deployed

(Courtesy of ASN)

Low Latency Routes

Today, the world’s oceans are spanned by in excess of 900,000km of operational subsea fibre optic cable systems, which carry over 95% of all international telecommunications traffic. This traffic comprises a mixture of voice, text, pictures, video and commercial data. By far the greatest concentration of this traffic is across the Atlantic between Europe and the Americas.

Advances made in system design capacity, combined with the ‘dotcom’ boom of the late 1990s, led to an unprecedented build of new transatlantic cable systems. The system owners were all chasing an over-optimistic forecast of massive growth in traffic. However, in reality, it was the supply of new capacity and not the demand for it that was spiralling. This meant that on transatlantic routes there was significant overcapacity, and this drove prices down. The low-capacity first-generation systems quickly became uneconomic to operate and were decommissioned. In addition, due to their inability to achieve predicted revenues, a number of the surviving second-generation optically-amplified systems changed hands at fire sale prices.

Although demand has grown steadily, the advances in transmission technology have enabled the existing systems to first absorb the traffic growth and then to be upgraded well beyond their original design capacity to accommodate further increase in demand.

Because of the large number of cable systems and the available capacity, the transatlantic market is one of the most competitive in the submarine cable telecommunications industry. Connections between the USA and Europe are critical to financial markets. In particular, fast connections between London and New York are important for a small number (15-20) of banks that engage in ‘high frequency trading’. For these companies, a few milliseconds difference in transmission rates can make a huge difference. They will always look for the quickest connection and if one bank has it, then the others must follow, so it is an all-or-nothing market where the customer is prepared to pay a significant premium. Therefore, in this segment of the market, achieving the lowest latency connection between principal global financial centres provides a competitive advantage to the cable owner.

Latency, or Round Trip Delay (RTD), is a measure of the time required to transmit a data packet from one location to another and back. Latency is a function of route length and system design. Hibernia Express currently offers the lowest RTD connection between London and New York.

With such large data-carrying capacity already available on both transatlantic and transpacific routes, the business cases that have emerged for the construction of new cable systems across both these oceans have largely been built on securing this low latency market.

Availability and Reliability

When purchasing capacity on the Atlantic route, for other than high frequency traders there are a number of aspects besides latency that will have to be taken into account. For these buyers, price will always be a primary consideration; however, availability/reliability will probably be the other significant decision factor. The fault history of a system will be a major concern, as disruption of the service, if too great, will require a high level of investment in restoration capacity on other transatlantic cable systems; but more importantly, interruptions to service could alienate customers who have the option of going elsewhere for their service. Across the Atlantic, the fault history of Apollo, supplied by ASN, stands head and shoulders above the competition, having had only one fault in ten years on each of its two cables. This is set against a global industry average of one fault every 2½ years for a single cable.

Industry Consolidation

In a further reorganisation of the telecommunications manufacturing industry in 2006, Alcatel merged with US company Lucent, and ASN became part of Alcatel-Lucent. Then in 2016 Alcatel-Lucent was acquired by Nokia. Nokia is now one of the world’s big four makers of equipment and systems for telecommunications networks. Two, including Nokia, are European, and two are Chinese. ASN is one of only three turnkey suppliers of submarine systems in the world; the others are NEC in Japan and TE Subcom in the USA, although NEC does not have its own in-house marine installation capability. Submarine fibre optic telecommunications cable is now manufactured only by ASN in France, Hexatronic in Sweden, Nexans in Norway, Norddeutsche SeekabelWerke (NSW) in Germany, OCC for NEC in Japan and TE Subcom in the USA. The major manufacturers of repeaters and BUs are ASN at Greenwich, NEC and Fujitsu in Japan, and TE Subcom in the USA.

For information about cable from the Telegraph Era  (Click Here)

For information about cable from the Telephone Era (Click Here)

A more detailed history of cable manufacturing on the Enderby Wharf site go to:

http://atlantic-cable.com/CableCos/EnderbysWharf/

Advertisements