Systems containing optical fiber have design lives on the order of decades so that models for assessing the mechanical reliability of the fiber must rely on extrapolations from accelerated short term testing. Such extrapolations are only valid if all relevant mechanisms are fully understood. The physical processes giving rise to mechanical degradation are reviewed and it is shown that no single model describes all situations. In particular, strong “pristine” fiber can behave quite differently from weaker fiber. Additionally, two degradation regimes are identified, one which is stress assisted (fatigue) and one which can occur even in the absence of applied stress (aging). Recent advances in understanding these phenomena are discussed and promising areas for future work are proposed.
This paper reviews the common techniques for mechanical testing of optical fiber specimens and compares and contrasts their attributes. Any technique must be able to grip the specimens without causing failure at the grips. The techniques generally fall into two categories; uniaxial tensile testing in which the fibers must be gripped carefully, and bending techniques which reduce the local stress at the grips. The former techniques generally give results that are easier to interpret due to the homogeneous stress field but are experimentally the least convenient. In contrast, bending techniques are experimentally convenient, but due to the inhomogeneous nature of the stress field developed and the short fiber test lengths, may not be as useful as tension for some purposes.
Optical fiber coatings, such as polymeric, metallic and inorganic are routinely applied on fibers. Polymeric coatings, particularly ultraviolet (UV)-cured acrylate are extensively used in the telecommunication industries. Metallic coatings, such as tin coated fibers showed strength values near to the theoretical value of pure silica. Inorganic coatings, such as amorphous carbon coated fibers showed comparable strength and high fatigue resistance for using these fibers as a potential long term reliable fibers. A review of different optical coatings, current status and future direction is presented.
It is known that hydrogen induced losses can occur in silica based optical fibers. There are several different mechanisms by which these losses can occur. Molecular H2 can cause losses in any silicate fiber, regardless of the glass composition. Hydrogen can also react at defect sites in fibers, giving rise to OH formation and other losses. The reaction rates and the effects on the loss of a fiber are dependent on the doping used in making the fibers. Hydrogen loss effects are reviewed for fibers of various types, including Erdoped amplifier fibers. The use of hermetic coatings as a means of avoiding hydrogen induced losses is briefly reviewed. Where possible, the methods used to predict long term loss increases are also discussed.
Although the service reliability of passive optical components has been quite good, methods for predicting reliability have not been developed for them as they have for fiber. There is a growing need for methods of predicting failure rates as these components move into new areas of existing networks and new applications where they will be subjected to a broad range of service environments. The methods will have to be based on accelerated tests derived from life-stress models appropriate for the known failure modes. This paper deals with the application of these basic principles to fiber optic connectors and splices. The common failure modes are reviewed and guidelines are proposed for developing reliability prediction methods.
This paper discusses the subject of the reliability of optical fiber splices. This is a subject of great importance to the telecommunications industry, which is already heavily dependent on fiber in inter-exchange and inter-office portions of the network and is beginning to invest in fiber in the distribution portion of the plant. As Fiber in the Loop systems  are deployed, the network will increasingly require splicing activity in construction, grooming, and maintenance. It will be necessary to be able to splice rapidly, cheaply, confidently (in difficult environments), and reliably in order to succeed in the telecommunications business of the future.
The fabrication, performance characteristics and reliability results of InGaAsP semiconductor lasers and photodiodes are described. Our results show that the lasers exhibit excellent light emitting lifetime and spectral stability and photodiodes exhibit excellent stability in dark current and responsivity for commercial lightwave system application.
The survivability for satellite applications of two classes of quantum-well-based fiberoptic light sources was evaluated by MeV-proton space-environment simulation studies. The first was an InGaAs/GaAs strained-layer quantum-well (QW) laser; the second was a broad-band light-emitting diode (LED) based on dual asymmetric quantum wells in the InGaAs/GaAs/AlGaAs system. In contrast to earlier reports comparing bulk active-region heterostructure LEDs with similarly structured laser diodes, these QW LEDs were more tolerant of proton irradiation (-3dB power at ~3xl013 protons/cm2) than the QW lasers (- 3dB power at ~3xl012 protons/cm2). The LEDs were operated far into gain saturation with a high-loss cavity structure, while the lasers were operated in a region where gain was more sensitive to current density. Therefore, atomic displacement-related recombination sites had a greater detrimental effect upon the lasers than the LEDs. The lasers held constant slope efficiency, and current thresholds increased linearly with proton fluence, while both LED power and slope efficiency decreased with proton fluence. The degradation was similar to that predicted from a universal damage relation for GaAs electronic devices, and extends that relation to include these QW photonic devices.
We review radiation effects in components and subsystems for fiber based satellite data bus architectures. These busses are designed for telemetry and command as well as payload applications spanning data rates from Mbps to Gbps. Issues include total ionizing dose to fibers resulting in increased attenuation, total dose and displacement damage in optoelectronic sources and detectors, and single particle transient effects in optoelectronic components. In each case we review component selection criteria and develop quantitative treatments based on experimental data to demonstrate how survivable data busses can be implemented even in severe orbits. We conclude that the many advantages of fiber based data links and busses will soon be made available for emerging satellite requirements to transmit data reliably at rates from a few Kbps into the Gbps regime.
Major factors in assuring long lifetime and fail-safe operation in optical communications networks are reviewed in this paper. Reliable functionality to design specifications, complexity of implementation, and cost are the most critical issues. As economics is the driving force to set the goals as well as priorities for the design, development, safe operation, and maintenance schedules of reliable networks, a balance is sought between the degree of reliability enhancement, cost, and acceptable outage of services. Protecting both the link and the network with high reliability components, hardware duplication, and diversity routing can ensure the best network availability. Case examples include both fiber optic and lasercom systems. Also, the state-of-the-art reliability of photonics in space environment is presented.