Industrial Measurements for Single Mode Fiber

Abstract

Industrial Measurements for Single Mode FiberRichard S. LoweNorthern Telecom Optical Systems Division8 Colonnade Road, Nepean, Ontario, Canada K2E 7M6AbstractThe purpose of this paper is to review our development of single mode fiber testingmethods for use in the manufacturing environment. The rationale for selecting particularmeasurement techniques will be discussed in terms of precision, speed, the relevance of theparameter determined and the ability to integrate the test equipment into an efficient andcost effective measuring station. In addition, the suitability of current industrialmeasurements for qualifying newer fiber designs, such as dispersion shifted fiber will beaddressed.IntroductionIn the rapid growth field of single mode optical fiber telecommunications there existsa need on the part of fiber /cable suppliers to provide a suitable specification of theirproduct in order to guarantee its quality to the customer. Over the past several years,considerable progress has been made in the laboratory towards defining what constitutes ameaningful parameter set for specifying conventional single mode fiber to the point that,today, standardization of a number of the current test methods has virtually been realizedthrough groups such as the CCITT, EIA and IEC. This is true for measurements of attenuation,cutoff wavelength and fiber geometry. The challenge for the test set designer in theseinstances has been to adapt and /or further develop existing test methods for use in themanufacturing environment in order to meet certain prescribed levels of precision, speedand accuracy, taking into consideration the skill of the operators performing the measure-ments.However, not all testing techniques make the transformation from the laboratory to theshop floor well. In these cases the designer may be required to develop a completely newtesting procedure specific to the factory environment. This has been particularly the casefor the measurement of chromatic dispersion and mode field diameter. For the former, theNd -YAG laser -based test method' is widely accepted as a reference technique; however, itssize, cost and complexity make it unsuitable for factory or field use. The result was thedevelopment of the phase shift technique2'3 utilizing several light emitting diodes orlasers to measure chromatic dispersion in the vicinity of the specified dispersion minimum.In the case of mode field diameter measurements, the transformation has been hampered notonly by the usual difficulties inherent in designing a factory test set but also byquestions raised regarding the exact definition of this parameter in terms of the raw dataacquired either in the near or far field. Until such questions are resolved, it may bepremature to talk about the factory test method.The test method development philosophy that is described below, revolves around measuringmeaningful single mode parameters quickly and accurately. Also, a serious attempt has beenmade to choose techniques that are compatible so that they can be integrated into a singletest station and thereby reduce equipment redundancy. Finally, the adaptability of the testmethod to the measurement of newer, more complex fiber designs without major equipmentchanges has figured in the selection process.A parameter set for industrial testingThe choice as to which measurements to perform in the manufacturing environment has beendetermined to a large extent by the customer's need for a functional product specification.ATTENUATION at the operating wavelength(s) clearly meets this requirement since thisparameter will influence the maximum repeater spacing achievable for a single mode trans-mission system. Measurements are currently needed at 1.3pm and 1.55pm; the latter to allowfor future upgradeability to take advantage of the lower intrinsic loss at this wavelength.Although fiber attenuation can be performed by the cutback technique4 using lasers orfiltered LED's at the appropriate wavelengths (this is in fact done at the cabling stage ofproduction) we chose to use a monochromator -based test set for three main reasons.