Make the connection - Technology Information

Communications News, Feb, 2001 by Jim Hayes

New `old' techniques simplify testing fiber-optic connectors.

Small form factor (SFF) fiber-optic connectors have already become widely accepted in fiber-optic networks. They are used in networking equipment to increase port density in hubs, switches and routers, with subsequent cost savings. Fiber-to-the-desk applications are adopting them for the high density at patch panels, easy termination and cost savings.

Some of these connectors, however, have major physical differences from the standard straight tip (ST) and snap-on connectors (SC), now widely used in fiber-optic LANs. Most require learning some new procedures for termination, testing and troubleshooting. The two procedures described here were actually developed more than 15 years ago for the first generation of fiber-optic connectors.

At least one of the SFF connectors--the LC--is a smaller version of the SC. One type of connector is really two ST ferrules in a common body, which makes it more conventional. The VF-45 and MT-RJ, however, are duplex connectors, in a plug-and-jack configuration.

Jacks are field-terminated, but patch cords use factory-made plugs, requiring terminating two fibers at once in the jack. The VF-45 uses a v-groove splice in a sealed enclosure to make the termination, using proprietary 3M tooling.

The MT-RJ jack has a prepolished fiber stub that is spliced simultaneously

with two fibers to complete the termination. The process involves stripping and cleaving the fibers, then inserting the cleaved fibers in the jack. Making a good termination requires having a good cleave and inserting the fibers fully before locking them in place.

Since the jacks are really just a mechanical splice, using a visual technique to confirm proper termination is possible. By inserting visible laser light from an instrument called a visual fault locator (VFL) into the jack with a patch cord, light lost can actually be seen in the splice portion of the jack. Of course, this cannot be done with a system source, which is in the infrared; the visible laser source is needed.

Some jacks terminate one fiber at a time, while others require two simultaneous terminations. The VFL must be used on both fibers at once if they are terminated together to get proper results. The VFL can also be used to find bad jacks if the back of the jack can be seen, since a bad jack will show visible loss.

Once the jack is terminated, it needs to be tested. The installed link to be tested is from jack to jack, so test equipment needs to be interfaced to that configuration. The big problem in making this test is setting a reference for "0 dB," or no loss. Industry standards, including EIA/TIA 568A, Annex H, and network standards require setting the 0 dB loss reference at the output of the launch cable, which then makes a measurement that includes the loss of the connectors at both ends of the cable plant.

The VF-45 and MT-RJ make this more difficult, since the connectors are hard to interface to some test instruments. Some power meter manufacturers offer adapters to these connectors, but the new breed of duplex cable testers have fixed ST or SC connectors. Adapting these instruments to test according to industry standards (with a one-cable reference) has been difficult until a technique from the early days of fiber optics was revived.

In the early 1980s, when fiber optics was new and testing standards were first being written, the cable test standard FOTP-171 was written to adapt instruments to all the connectors. This standard includes an option for using what are called bucket fibers, or large core fibers on the meter. These large fibers act just like detectors, gathering all the light from the end of the fiber exactly like a large area detector.

By creating a bucket fiber cable with a SFF connector on one end and two STs on the other end, the test instruments can be adapted with fixed ST connectors to use the one-cable reference required by the standards. The bucket fiber cable is then left on the meter end to make all the measurements of cable plant loss. If the instruments are bidirectional, the adapter cable must have the bucket fiber for the meter input and a regular fiber for the source output at the instrument.

Several manufacturers of the MT-RJ have already adopted these two simple procedures to simplify installation and testing of their connectors. As more experience is gained with these connectors, more innovative ways to work with them will be found, simplifying the processes.

Hayes is founder and president of Fotec, Medford, MA.

www.fotec.com

Circle 260 for more information from Fotec