Multimode- vs. singlemode-fiber standards
By WILLIAM B. GARDNER
The first optical-fiber standards, written in the late 1970s, dealt exclusively with multimode fibers. When singlemode fibers entered the marketplace in the 1980s, there was a natural desire to carry multimode concepts, definitions, and test methods over into the singlemode standards. Sometimes this worked (as with cladding diameter and proof testing, for example), but many times it didn`t. Here are three examples of multimode concepts that do not apply to singlemode fiber:
Numerical aperture (NA)--Multimode standards such as the International Telecommunication Union`s (itu`s) Recommendation G.651 on 50/125-micron graded-index fiber define a maximum refractive index of the core and a cladding refractive index. These two parameters are then used to calculate the theoretical NA of the fiber. The actual NA is determined by measuring the divergence angle of the cone of light leaving the fiber (see, for example, the Telecommunications Industry Association [tia] test method fotp-177). The theoretical and actual NA are nearly equal and provide a useful measure of how much light the fiber can capture from an incoherent source.
The concept of NA is based on geometrical optics, or "ray-tracing," which works well in multimode fibers but not in singlemode fibers. All of the itu`s singlemode recommendations (G.652, G.653, G.654, and G.655) contain the words "the refractive index profile generally does not need to be known." Although singlemode profile measurements are difficult because of the small core, they can be done. But even if the maximum refractive index of the core were measured, inserting that value into the formula for theoretical NA would not give the correct divergence angle for the singlemode fiber`s radiation cone.
Because NA is strictly a geometrical optics concept, the term isn`t even defined in singlemode fiber standards.
Core diameter--The automated gray-scale analysis, which is so useful in making accurate multimode core diameter measurements (see tia`s fotp-176), is of little value in singlemode-fiber measurement, because the core size is so close to the resolution limit of optical measurements. This fundamental resolution problem contributed to the replacement of core diameter with mode field diameter in singlemode fiber standards. Core diameter and NA are important for predicting splice loss in multimode fibers; mode field diameter alone serves this purpose in singlemode fiber.
Prescribed launch conditions--Because attenuation and bandwidth values in multimode fibers depend on which modes are excited, the test methods prescribe how the launch conditions are to be controlled. For example, tia`s fotp-50 offers two options for attenuation measurements: mode filters and beam optics. The mode filter is created by wrapping five turns of the fiber around a mandrel whose diameter is 20 to 25 mm (the exact value depends upon the fiber type).
There is no such requirement for selecting the modes to be launched when measuring singlemode fiber attenuation, since the fiber propagates only one mode. Nevertheless, the use of short (a few meters) fiber lengths in the measurement could allow the presence of a higher-order mode to influence the measurement. tia`s fotp-77 provides guidance on the selection of a mode filter (yes, a mandrel wrap) to ensure that this doesn`t happen. But the considerations are not the same as those in fotp-50.
Equivalences between multimode and singlemode fiber concepts are not only imperfect, but sometimes even nonexistent. q