Published In October 2000
In the past, whenever I trained a group of technicians about fiber testing, I would ask if anyone knew what OTDR stood for. Next, I would rattle-off “optical time domain reflectometer,” then jokingly say, “You must memorize that before you can pass the course.” Many technicians in the voice/data field today can call my shot because they have heard, seen, read about, or used an OTDR. Although this test instrument is becoming commonplace, it may not be used correctly, or to its full capacity. This article is meant to shed light on features and functions of this piece of test equipment with a name so long we’re thankful for acronyms. Applications Today, many fiber optic cable installation contracts specify that the cable must be tested before and after installation. Cable pretesting can be performed using an OTDR. All fiber optic cable that is to be used for a job should be pretested prior to installation. This protects all parties involved, from the manufacturer to the installer. The OTDR is a “near-end-device,” which means nothing is required at the far-end, so technicians like using it for cable pretesting. Nearly all outdoor cable installations, and many indoor cable installations, require OTDR testing as part of the acceptance testing procedure. The OTDR performs an “insertion loss test” as defined by the EIA/TIA. Fiber type and wavelengths Before you perform an insertion loss test, you must ensure you are using the correct OTDR source type for the fiber being tested. If you are testing multimode fiber, you must use a multimode module, and if it is single-mode you must use a single-mode module. The EIA/TIA specifies that you test at least two wavelengths in each category. For multimode, 850 and 1300 nanometers is the most common; for single-mode, 1310 and 1550 nanometers. The wavelengths are all infrared, and are invisible to the naked eye. Remember, individual specifications could require testing of other less-common wavelengths, which could get very expensive. Analyzing events The OTDR exhibits the information that it acquires onto an LCD or a CRT. The graphic image that is displayed may have one of the following names, depending on which manufacturer and whether you are dealing with a “trace” or “acquisition.” (Refer to Figure 1.) I will use the term “trace” for this discussion. An “event” is any anomaly or abrupt change in the OTDR trace. Events are usually either a splice, bend, or break in the fiber. The OTDR gives us three key pieces of information about an event: loss, distance, and back reflection. The loss, back reflection, and overall distance limits are subject to the test parameters for a particular project. A positive spike or pulse in the trace is back-reflection, and a drop in the trace is loss. The higher the spike is, the more back-reflection. The greater the drop, the more loss. OTDRs can experience “pseudo-gain” or a “gainer.” This usually occurs when two joined fibers are poorly matched. A gainer appears as an abrupt up pulse and remains steady until the next event. In most cases, the gain will be loss when the fiber is “shot” in the reverse direction. The gradual downward slope of the trace is the loss of the fiber. The “noise,” or “grass” at the end of a trace is sometimes mistakenly called “backscatter”; which is the usable portion of the trace. Remember that the end of the fiber is also an event. Loss and distance scales Almost all OTDRs have loss and distance scales, which are to be used for a reference. The loss scale is located on the left side of the screen. Remember that this scale is relative, because you can move the trace up and down on the screen. The distance scale is located on the bottom of the screen, and lets you keep track of your range. You need to keep an eye on both of these scales because as range and loss settings change, so will the size and shape of the trace. Don’t jump to conclusions by the size of events on the screen; read the numbers! Range Setting the range properly is critical because too short of a range will not let you see the entire length of fiber, and may appear as a broken fiber. If you do not know the length of the fiber, start with a range that should be much longer than the fiber being looked at, so you can see the entire length. After you can see the end, adjust the OTDR to the shortest range available that still exceeds the total length to give you optimal horizontal resolution. Some OTDRs allow setting the range on a short range, such as 50 meters and you can scroll down the fiber looking for hard-to-find events. Index of refraction The index of refraction must be set to match the fiber being tested. The index of refraction is the ratio between the speed of light in a vacuum, and the speed of the light in the fiber’s core. It is a number such as 1.468. The inverse of this number will give you the percentage at which the light travels in the fiber’s core compared to that in a vacuum. An index of 1.468 is just over 68 percent. In most cases, we don’t communicate at the “speed of light,” usually around .66 of a second. If this setting doesn’t match that of the fiber, it will give you a false distance measurement. In reality, due to other measurement obstacles, on distances less than 10 kilometers the default value will get you close enough. For event locating, be sure to set your OTDR to the same setting that was used when the records were created. Pulse width The pulse width is another parameter that can be set by the user. The pulse width is the length of time that the source is left on or “fired.” Pulse widths are measured in nanoseconds (billionths of a second). Longer pulse widths have more energy, and can travel further, thus we use a long pulse width on a long piece of fiber. Short pulse widths have less energy and will not travel as far. So why not use a long pulse width all the time? Long pulse widths give less resolution and create larger dead zones. If the pulse width is too long for the length of fiber, you could pass up an otherwise visible event. If the pulse width is too short for the length of fiber being tested, then it will appear as if you have a broken fiber. So, there is no free lunch. The range and the pulse width need to be adjusted accordingly as you fully examine the fiber. Averages The number of averages determines how “clean” the trace will be. Short lengths of fiber require fewer averages than long ones do. The number of averages can be from a few hundred to over a million. The final trace is an average of many “shots,” not just one. By averaging, we can get more consistent repeat testing. The more averages that are taken, the more time it takes to test. Some “shots” take only a few seconds, and some could take over an hour. Enough averages must be taken to clean up the trace enough to make all of the events visible. Dead zone Dead zone is an area of the OTDR trace where no usable test data can be acquired. It is measured as a distance. The most popular dead zone is the initial one that occurs where the launch jumper is plugged into the OTDR itself. This initial dead zone is usually between 10 and 50 meters wide, depending on the type of OTDR and the settings used. Dead zones are also produced in events called “event dead zones.” Using a “pulse suppression jumper” can minimize the initial dead zone. This is a long jumper, usually 50 to 100 meters long, that allows the OTDR time to recover from the radical event of entering the fiber. You must keep in mind that an event dead zone will be created where the pulse suppression jumper is connected to the patch panel; this one is usually smaller than the initial dead zone. Also, if using a pulse suppression jumper, you must remember to subtract its length to locate cable events accurately. The length of the event dead zones govern how far apart events must be for both to be detected. For instance, two splices placed 10 feet apart are nearly impossible to analyze. The dead zones of field-use OTDRs have been greatly reduced over the past 10 years, through better hardware and software. Test result documentation Testing is a waste of time and money, if records are not kept of the work that has been done. Modern OTDRs have internal memory for storing records. Most testing specifications require contractors to provide a hardcopy and a floppy disk of the test results. Contractors should keep their own records for liability and troubleshooting purposes. Producing records can be time consuming and expensive, so don’t omit test results from a quote. Light at the end of the tunnel After learning about all the factors that may affect the readings, and all of the adjustments that must be made, you are probably exhausted. We used to make all of these adjustments manually, and use our experience to acquire the necessary data. Today, almost all field-use OTDRs make these adjustments automatically and give you the loss, back reflection, and distance measurements in a chart form. So why bother? Some test specifications require you to set the OTDR to prescribed parameters. Another important application is troubleshooting; a well-trained and experienced technician can find problems that a “one-button operator” can’t. The automatic testing modes on the OTDRs work very well. But, we still need to know what the numbers mean and how to dig a little deeper if necessary. If you purchase or lease an OTDR, practice with it before you take it on the job site. You will need a few hundred meters of fiber to connect to; it isn’t good for an OTDR to “dry fire.” If you practice on already-installed fiber, do not connect it to an active fiber; this could damage the OTDR or the communications equipment. Don’t forget about the hazards that lasers can impose. NORRIS is training director of Midwest Telecom Training, LLC. He can be reached at (812) 254-3488 or firstname.lastname@example.org.