Fiber mechanical reliability test update
HARISH C. CHANDAN
The mechanical reliability of optical fibers is the focus of the effort of Telecommunications Industry Association (tia) Working Group FO-6.6.8. Chaired by Hakan Yuce of Bell Communications Research (Bellcore), this group is responsible for developing test procedures for fiber reliability parameters. Working Group FO-6.6.8 is instrumental in establishing the U.S. position on various test procedures presented to the International Electrotechnical Commission (iec) and the International Telecommunication Union (itu).
Since the last review of fiber strength and fatigue in Lightwave (see September 1993, page 30), significant changes have occurred in the test procedures, and some specifications for reliability parameters are under review. The revised Fiber Optic Test Procedure (fotp) for prooftest ( tia-fotp-31c) was issued in February 1995. The old fotp-31b required the fiber under test to endure the specified prooftest load for one second. If the load duration was less than one second, fotp-31b prescribed the use of a higher prooftest load value. This value was designated as being a small percentage higher than the specified prooftest load.
However, this test procedure ignored crack growth during unloading, which could reduce the strength of a prooftest level flaw to a value less than the prooftest level. Even though such flaws are rare, discussions in FO-6.6.8 led to a change in the prooftest procedure: It no longer requires a one- second dwell time at the prooftest load. Instead, it allows for a correction term based on the unloading time.
With this correction, the prooftest load becomes a small percentage higher than the specified prooftest level to ensure that the minimum strength after prooftesting is equal to or higher than the specified prooftest level. This change from the one- second dwell time to minimum strength after unloading is going through approval at the iec and itu.
An effort is under way in the tia to combine the strength (fotp-28b) and dynamic fatigue (fotp-76) tests to create a single test (fotp-28c). Discussions are being held about the number of stress rates to be used and how far apart they should be.
Some fotps under review include the measurement procedure for static fatigue by axial tension (fotp-117), dynamic strength of optical fibers by two-point bending (fotp-90), dynamic fatigue parameters of optical fibers by two-point bending (fotp-116), and the static fatigue of optical fibers in two-point bending (fotp-97).
The discussions and round-robin testing in FO-6.6.8 are helping to redefine some of the existing industry specifications to make them more meaningful. One such specification, which is in the process of being changed, is for the Weibull m, which is the slope of the failure probability versus strength plot. The Weibull m is inversely related to the standard deviation in the strength measurements. A high Weibull m value means that the strength distribution is very concentrated, whereas a low value means that the strength distribution is spread out.
Based on the FO-6.6.8 discussions, Bellcore published an issues list report in December 1996 (Generic Requirements, gr-20-ilr, Issue 1B). This report includes a proposal for changes in the specification of Weibull m and other reliability parameters. The Bellcore proposal suggests replacing all criteria involving the Weibull m with criteria involving strength at the 15% Weibull probability level. The FO-6.6.8 discussions revealed that the value of m is only an indirect indicator of fiber performance for both unaged and aged samples, but it is somewhat related to the manufacturing process. Importance was given to the lower-probability values of strength that were more closely related to issues such as handleability, which is the ability of fiber to be handled during splicing and routine measurement operations. q