As data center network gradually migrates to high-speed 40G and 100G Ethernet, many IT managers prefer to deploy MPO/MTP based fiber cabling for its easy handling (MPO cable ispre-terminated in the factory) and high density performance, but since the structure of MPO/MTP is more complicated than standard fiber optic cabling, it is critical to have proper maintenance procedures to keep the network working in good condition. For MPO/MTP cabling, testing is one of the most important process in routine maintenance. This post will introduce how to properly testing MPO/MTP cabling.
Pre-terminated MPO/MTP cabling is only guaranteed good as it exists in the manufacture’s factory, but after transported, stored, and later bent and pulled during installation in the data center, all its performance becomes uncertain, so proper testing after installation is the only way to guarantee performance in a live application. Besides, the use of MPO cables for trunking 10G connections in the data center has steadily risen over the past 10 years, and now data center are gradually upgrade to 40G and 100G, which means each time you migrate, you need to verify the links to ensure the performance delivery the organization requires.
As long as testing is necessary, how can we operate the proper MTP cabling testing? Generally, testing a parallel cable plant with MPO connectors is once called impossible in an industry meeting. The problem is that few test sets have adapters for these multifiber array connectors, so you need to find ways to adapt typical single fiber test equipment to testing them. In doing so, you must compromise the typical insertion loss test by adding adapter cables from the MPO connector to LC or SC type connectors more commonly used on test equipment.
Solution 1: Using a typical single fiber optic test source and power meter. In this case, you have to add MPO LC breakout cable to both MPO ends of the cable plant and then test the cable plant including the breakouts. In making the measurement this way, your loss will be higher than the actual cable plant by the loss of the connectors on the breakouts when connected to the launch and receive reference cables as shown below. This will add approximately 0.5-0.8dB +/-0.2dB to the actual loss of the cable plant under the test. You can compensate for this by using 3-cable reference (as shown in the table below) but will increase uncertainty.
Solution 2: Using a single output source to connect each fiber in the cable and a meter with a large area detector and MPO adapter. You manually change the source from the fiber to fiber, but the meter remains fixed as shown below. This method will not confirm the polarity of the connection unless your carefully note the fiber and the source are connected to each time.
Solution 3: Using a source with breakout and a meter with a large area detector or a breakout and a single fiber meter. You switch the source from fiber to fiber using the switch which has relatively low variability from fiber to fiber as shown below.
Solution 4: There are testers available with MPO interface designed to test these cables efficiently. Although these testers are expensive, White these testers are expensive, they are probably the best way to test these cables. The main procedures are shown below.
As more and more MPO/MTP based fiber optic cabling used in data centers, it is necessary to do routine testing after network upgrading or installation. This post has introduced four MPO/MTP testing solutions. Hope you can useful information from it.