FOA Guide

Topic: Test Fiber Optic Cables With Different Connectors

Table of Contents: The FOA Reference Guide

Testing Cables With Various Fiber Optic Connectors

The most popular fiber optic connectors over the years (e.g. ST, FC, SC, LC)  have used a cylindrical ferrule to hold and align the fiber and have an mating adapter to align two identical connectors for mating. A securing device like a screw-on nut, bayonet latch or snap-in latch completes the design. This describes the majority of fiber optic connectors that have become widely accepted, like the SMA, ST, SC and the new small LC.
connecting 2 cables ST connectors and mating adapter
connector ferrule
All these connectors are easy to test. The protruding ferrule makes it simple to connect them to test equipment as well as each other. Testing loss was a two-step process: use a power meter to measure the power out of a reference cable with that style of connector on the end to establish the power launched into the connector being tested (one cable reference above), then attach the connector to test to that reference cable and measure the power transmitted through the mated pair of connectors and the rest of the cable.

Single-ended testing   double-ended testing

If you were testing a short patch cord using FOTP-171, this test only tested the connector mated to the reference connector, as a short length of fiber adds negligible loss. That’s just what you want, as you could test each connector separately and confirm it was good. Since it tests only one end of the cable, we called it the “single-ended” test. If you wanted to test an installed cable plant, you included a second cable at the power meter end, so you added the loss of the connector on that end of the cable to your measurement, as well as everything in between. We called this a “double-ended” test per OFSTP-14.
But what happens if your connector is not compatible with your power meter and you can’t  mate the launch reference connector to your power meter to set the reference? Or even worse, if the connectors are a plug and socket design like the MT-RJ, VF-45 or OptiJack and can only mate with their counterparts? They have jacks on each end of the installed cable plant you want to test and use patchcords with plugs on either end.

plug-jack connector  MT-RJ plug (L) and jack (R)

In the earliest days of fiber optics, resourceful engineers devised several ways to work around these issues by using hybrid fiber optic cables, a cable with one type of connector on one end and a different type on the other end and/or hybrid mating adapters. You must choose reference cables with connectors that mate to your test equipment on one end and the connectors on the cables you want to test on the other end. Then you can connect your cable to test in the middle to make the measurement just like other double-ended tests.

hybrid test cablesUsing hybrid test cables

hybrid mating adapters Hybrid mating adapters

The hybrid reference cable method works for almost any type of connector or test equipment, which is amazing with all the connectors available today and all the configurations of test equipment. The problem is deciding how to set your reference power for “0 dB” loss, especially if your power meter cannot accept the connector type being tested. Standards allow you to set the reference with one, two or three reference cables. The only caveat: you must document the method in your test data because it directly affects the measured values. (Table 1)

Table 1. Reference methods for fiber optic loss testing.

Reference Method
Reference Cables Connectors Included in Reference Measurement Estimated reduction in measured loss Estimated increase in errors
1-Cable Method 1, launch 0 0 dB 0 dB
2-Cable Method 2, launch and receive 1 0.2-1 dB +/-0.2 dB
3-Cable Method 3, launch, receive and “golden cable” 2 0.3-1.5 dB +/-0.25 dB

Here is a complete rundown on all standard methods of testing fiber optic cables.

Here are the FOA Standards for testing fiber optic cables.

If you set your reference   0 dB loss with just the launch cable (Method B), you set the reference at the output level of the launch cable, so any losses associated to cables under test – the mated connectors and any losses in the cable or cable plant itself – are included in the loss measured by the power meter.
If you set your reference with two cables, launch and receive (Method A), one connector loss – the loss of the mated connectors on the reference cables – is included in the reference. When you make a measurement of a cable you want tested, the measured loss will be lower by this connector loss, which is not really known. It could be as little as 0.2 dB or as much as 0.75 dB in the worst case. Our lab tests have shown the error of the measurement is higher also, by as much as +/-0.25 dB, due to the unknown connector loss included in the reference.

test plug-jack connectors

There is usually only one practical solution for the plug and jack type of connector. The three cable reference method (Method C) uses launch and receive cables, plus a known good cable like the cable you want to test (known good quality but only a short length of fiber) between them.  If you use the 3 cable reference, you include two unknown connectors, reducing all measurements by these connectors’ losses, from 0.4 to 1.5 dB, and the uncertainty can be over +/-0.25 dB.

Choose Your Standard!

  Setting the reference power is not a trivial issue.  US and international standards allow setting the “0 dB” loss reference with either one, two or three cables, as long as you document the method in your test results. The TIA-568 structured cabling standard says you should use Method B, the one cable reference. The  equivalent international standard (IEC/ISO 11801) says you should use the three cable method, as it is the only method that works with any style of connector. Communications network standards are no better, as they have used all three methods in their documents to define what cable plant loss is acceptable for their network.
The TIA-568 document has caused much confusion, as some structured cabling systems use connectors like the MTP or MT-RJ that cannot be tested by Method B, the one cable reference method, especially with test sets that commonly have fixed SC interfaces. Manufacturers of those test sets have come up with some of the most confusing kludges imaginable to test these connectors, including some that use several hybrid cables added after setting the reference for 0 dB loss. The potential error is very high. It’s probably one of the reasons that connectors like the MT-RJ have had some problems in the field – practically nobody can really test their loss.
Is there a solution? We recommend choosing the simplest method that works. (Table 2) For the usual 2.5 mm ferrule ST and SC, use Method B (1 cable reference) and for the plug and jack duplex connectors like the MT-RJ, VF-45 and OptiJack, use the Method C (3 cable reference.) But do keep your reference cables in good condition, clean them between measurements and document your methods for your customers. Sometimes the method used is less important than consistent test methods, as bad fibers and cables should stand out from the good ones no matter what the test method.

Table 2. Recommended reference methods for connectors

Connector Type Mates to test equipment Does not mate to test equipment
Plug to Plug with mating adapter Method A, B or C Method A or C*
Plug to Jack Method A or C Method C
Method A can be used if  the connector under test can be adapted to the connector interface on the test set – e.g. a tester with SC interface, but ST and FC can be mated to SC with a hybrid mating adapter, so SC reference cables can be used.

Here is a complete rundown on all standard methods of testing fiber optic cables.

Here are the FOA Standards for testing fiber optic cables.

(C)1999-2012, The Fiber Optic Association, Inc.