Get in Line

Artist's depiction of fiber cladding diameter.

Source: Corning Inc.

Sometimes little things mean a lot.

For example, microscopic changes in glass geometry can lead to substantial reductions in installation expense.

Splicing fiber accounts for between 3 percent and 5 percent of cost for metro/access networks. But as fiber moves deeper into the network with architectures such as fiber to the curb (FTTC), this figure can rise to nearly 10 percent (see pie chart, "First Installed Fiber to the Curb Cost--Single-Fiber Splicing"). That's because these networks typically use higher fiber-count cables that require more splices per kilometer (km), which translates into increased fiber handling and splicing by more people, and potentially higher installation costs.

Because splicing fiber is labor-intensive, it is important to do the job correctly the first time. Successful splicing depends greatly on the two spliced fibers being straight, of the same size and having their cores precisely in the center. The better the match between the two fibers, the less light will be lost at the joint. Industry modeling shows that improved fiber geometry can save approximately $1.20 per splice, or 6 percent of the typical $20 labor cost associated with single fiber splicing.

Fiber to the Curb Cost--Single-Fiber Splicing

Source: Corning Inc.

Improved fiber geometry also has promoted the development of mass-fusion splicing techniques for telecommunication. Mass-fusion splicing allows installers to splice many fibers at once, which can reduce splicing labor costs by as much as 60 percent.

The prospect of such significant savings is especially important to the rapidly expanding competitive local exchange carrier (CLEC) market. During a CLEC Network Deployment Forum hosted by Corning Inc. in early 1998, more than 40 percent of the 57 attending CLEC representatives indicated they would build networks in one to five cities in 1999. Ten percent of the respondents said they planned to build in 21 or more cities. The ability to deliver advanced communication systems quickly and economically will be key to CLEC success. Although fiber geometry specifications may seem like mere technical details, a strong understanding of fiber geometry and new splicing techniques clearly provide network installers a competitive advantage.

Fiber splicing is considered labor-intensive because it usually involves splicing fibers one at a time, except in situations in which mass splicing is used. In splicing, technicians typically break open a cable where the fiber is positioned in a tube, match each fiber by code to its mate and splice each pair individually. Good splice yields rely heavily on individual fiber-to-fiber alignment. As fiber becomes the backbone of CLEC networks, the number of fibers in a cable increases and so does splicing cost. Improvements in fiber geometry, however, have allowed many splicing equipment manufacturers to develop less-expensive fusion splicing equipment that relies on the physical alignment of optical fibers to optimize the splice rather than on more expensive computer-controlled optics.

Even so, single-fiber splicing remains a time-consuming and costly element for many network installers, so efforts to improve fiber geometry tolerances continue. These glass geometry improvements can deliver substantial cost savings to network installers (see chart, "Improved Fiber Geometry Cost Savings"), and tighter geometry has helped the industry develop better splicing technology. In addition, equipment is becoming less complicated and less expensive. Over time installers can expect overall network installation costs to drop through a reduction in capital investment costs, equipment maintenance expense, training requirements and labor efficiency.

Of course, splices are permanent connections between two or more fibers. It is here that tighter geometry tolerances become important. The glass geometry of an optical fiber affects how well fibers splice together and can impact optical performance.

Glass geometry is known to be a primary contributor to splice loss and splice yield--a fact that has been verified by industry studies and collaborative efforts among vendors. The collaborative confirmed that improving the tolerances of fiber geometry can significantly reduce splice loss and improve splice yields. Bellcore, the former research and development arm of the Bell companies that now does work for the telecommunications industry at large, also has acknowledged the efficiencies gained in splicing fiber with improved fiber geometry. As a result, Bellcore has set tighter fiber geometry objectives for fiber manufacturers.

The following improvements can lead to higher splice yields, lower costs and better overall system performance:

Fiber Curl: Curl is the amount of curvature along a specified length of fiber. Fiber has a degree of curvature that's precisely measured and expressed as "radius of curvature." Controlling fiber curl is a relatively new focus for fiber manufacturers, brought about by the increased use of mass-fusion splicing technology. Fiber curl is caused by the composition of the glass, thermal stresses and uneven cooling--all of which must be tightly controlled during the fiber manufacturing process. For mass-fusion splicing, a tighter curl specification equates to better splicing results.

Core/Clad Concentricity: Core/ clad concentricity refers to how well the core is centered in the cladding glass. Studies have shown that core/clad concentricity is the most significant contributor to splice loss. When cores are misaligned, light escapes. Tighter core/clad concentricity tolerances ensure that the fiber core is centered within the fiber diameter. This reduces the chance of core misalignment (or offset) when splicing two fibers together. The improvement in core/clad concentricity yields a better, lower-loss splice. In a mass-fusion setup, there is no opportunity to "tune" the fiber alignment (as can be done when splicing single-fiber), so the fiber's geometry is of primary importance in achieving acceptable splice losses.

Cladding Diameter: Cladding diameter is the outside diameter of the cladding glass region of a fiber. Tighter tolerances ensure that fibers are of similar sizes, and when the diameter is inconsistent, the fibers do not align properly during splicing.

Critical Mass

The CLEC market is the fastest-growing segment for fiber deployment in North America. It represents more than 10 percent of the North American market. There are established CLEC networks in 47 states and 219 cities. Fiber deployment in the CLEC market grew by 75 percent in 1995, 68 percent in 1996 and 50 percent last year. It is estimated that the number of CLECs could increase from 50 to 500 during the next eight years. Current and future market dynamics require high-performance backbone networks. To meet that need, CLECs continue to demand easier splicing, lower losses, increased ruggedness and better handling of the fiber they choose to deploy.

The deployment of fiber deeper into the network has fueled the growing use of ribbon cable among CLECs in the United States. Ribbon fibers are connected to form a flat, ribbon-like structure within a cable, and are encapsulated in a polymer matrix material. Typical ribbons contain four to 12 fibers, and up to 12 ribbons can be stacked together to form the core of a cable. The simple structure of a ribbon fiber makes it easy to splice in the field--reducing splicing time and cutting labor costs.

With the use of ribbon cable, the advent of mass-fusion splicing technology promises even greater savings for CLEC network installers and owners. The whole idea of mass-fusion splicing is speed, as installers do not want to have to redo mass-fusion splices, and the goal is to achieve acceptable losses the first time. In a mass-fusion splicer, 12 fibers are aligned in v-grooves and at once. The fibers must be as straight as possible to achieve the best splicing results. Because this method does not allow technicians to manipulate fibers individually, tight geometry tolerances are essential. Mass-fusion splicing has generated great interest in the industry because of the cost savings potential. Mass splicing enables faster splice time, which results in less labor and, therefore, lower labor costs. In the United States, the cost savings from mass splicing can be very significant, depending upon the distance between splices. For example, if splices are located every 2.5km and 12-fiber mass splicing is performed, installers can save about $10 per installed fiber km, which could save $144,000 for a 100km, 144 fiber system. While the adoption of mass-fusion splicing in the telecommunications industry is just beginning, installers are finding it may be a viable alternative to single-fiber splicing. The key to this alternative, however, is high-quality fiber.

The same is true for blind splicing, another method for reducing splicing costs. Fiber with tightly controlled geometry tolerances allows installers to reduce the amount of splice testing. Instead of measuring splice performance at each splice point, one measurement is performed on the completed system link to ensure that total loss is within the system budget. By eliminating the individual site measurements, an installer can save approximately $5 per splice, which translates into thousands of dollars in overall system costs.

To advance in the competitive carrier marketplace, CLECs must maintain a solid understanding of technological advancements and various deployment strategies. As today's networks scramble to adapt to the rapid shift in consumer demand from voice to more data types of services, fiber geometry specifications become less of a technical detail and more of a competitive advantage.

Jane Li is the competitive local exchange carrier (CLEC) market manager for Corning Inc., Corning, N.Y. She can be reached via e-mail at lij@corning.com.