Premises fiber optic cabling link testing requires precise referencing methods to obtain accurate and valid results. Loss testing for multimode fiber cabling is specified in ANSI/TIA/EIA-526-14A. This standard contains two test procedures: Method A and Method B. This article describes both methods, and explains why Method B is the proper one for testing fiber links contained in premises networks. This article also proposes a new test procedure as an adaptation to Method B. This new test procedure is the preferred method because it provides results conforming to Method B while offering installers more flexibility for testing fiber links with all types of connectors, including small form factor (SFF) connectors. Other advantages of the Method B adaptation for simplifying the testing process and reducing the opportunity for errors are also described. Method A Method A is used for testing links in which the loss in the fiber cable dominates the total attenuation, rather than the loss of the connectors, as is often the case for telecommunications networks. Method A’s referencing procedure uses two patch cords and an adapter connector for each fiber link to be tested. The two patch cords and one adapter connection have been referenced out when the test is performed. The test results include the loss of the fiber link under test plus only one connector. While this method has been used effectively to test long-haul telecommunications fiber links, it is not precise enough for today’s premises market. Method A includes only one connection, but the network operation actually experiences the loss of the fiber link plus the connection at both ends. Consequently, Method A understates the power loss in the link. This is not an issue for long-haul telecommunications links, since most of the loss is in the long lengths of fiber, with minimal loss in the precision connectors. However, in premises applications, fiber lengths are very short and the amount of loss in the fiber cable itself is minimal. The majority of power loss is found in the connections at either end. The increasingly stringent power loss requirements of applications like Gigabit Ethernet require the entire link loss to be measured. That is where Method B becomes applicable. Method B Method B is used for testing links, such as premises links, for which the connector loss is a significant portion of the total attenuation. The referencing procedure for Method B uses one patch cord per fiber link to be tested. Since only one patch cord (per link) is part of the reference, the test results will include loss from the fiber cable under test plus the connectors at both ends. (See blue section in Figure 4.) Technically, it will also include any loss in the additional patch cord, which is negligible because the length is so short. For premises fiber networks, this method provides an accurate measure of loss in the fiber link because it includes the fiber cable plus the connections at both ends. However, when using Method B, be aware of the following shortcomings: When going from the reference setup to the test setup, it is necessary to disconnect one end of the patch cords from the tester. It is very important to never disturb the connection at the “output” or source end, because if it is, the reference is lost, and proceeding without re-referencing will seriously compromise the test results. Unfortunately, one could easily disconnect the patch cord from the source (output) end instead of from the detector or (input) end. Although you must disconnect the patch cords from the detector (input) end of the tester, extreme care is required, because dirt and other elements can damage the detector. To test SFF connectors that contain both transmit and receive fibers in the same connector, you are forced to disconnect from the source (output) end in violation of proper referencing and test procedures. If you use Method B, it requires that you have the same type of connector on the tester, as you will be testing in the fiber link. The following section describes a new test procedure that provides the same test results as Method B and preserves integrity to testing standards, while overcoming its shortcomings. Adaptation to Method B A simple adaptation to Method B allows us to retain the accuracy (every measurement includes the cable and both connectors) while avoiding the major disadvantages. The referencing procedure for this adaptation is performed using two patch cords and an adapter connector per fiber link to be tested. However, the test procedure is new. The test procedure includes the addition of a short test jumper with a connector so that the test results will be the same as those obtained with Method B. Just as Method B’s results, these contain the loss for the fiber cable plus the connectors on both ends. The two patch cords and connector per link from the reference setup have been referenced out. Make the correct loss measurements The Method B adaptation offers several key advantages over the original Method B while preserving its accuracy. The Method B adaptation’s loss results conform to ANSI/TIA/EIA-526-14A, Method B. According to Method B, to measure the link loss correctly, the test path must have two more adapters in each fiber link than in the Set Reference path. The test procedure described in this article adheres precisely to this requirement. In this way, the measured loss will be the loss of the fiber in a link plus the loss of a connector at each end of the link. This value of loss is the real value network application hardware encounters. The Method B adaptation allows the use of hybrid patch cables to connect test equipment to the links being tested. This allows consistent testing of links with all types of connectors, including those that use small form factor (SFF) connectors. Preserve the integrity of your test The adaptation to Method B makes it unnecessary to disconnect the patch cords from the test equipment, thereby reducing the possibility of errors caused by reinsertion of patch cords or by contamination or damage of test equipment fiber interfaces. Take the complexity out of testing SFF connectors While single-fiber sources and power meters have been used effectively for testing fiber links containing single-fiber connectors, testing dual-fiber connectors with single-fiber testers is awkward and prone to errors. Dual-fiber testers are the best way to test duplex connectors and therefore are recommended for SFF testing. TAYLOR is manager for Fluke Networks’ fiber optic test products business. For more technical information, contact Fluke Networks at 1 (888) 99-FLUKE or visit their Web site at www.flukenetworks.com.