Portable OTDRs Simplify Complex Fibre Tests

Read Test & Measurement Europe's Survey of Portable OTDRs

The concept of OTDR operation is simple in that the unit shoots a pulse of laser light into the end of a fibre and then looks for reflections. The return time and the nature of the reflections characterises the fibre and locates any discontinuities along its length. Reflections from general light backscatter determine the underlying attenuation along the fibre. Particular reflections from discontinuities (called “events”) identify connectors or splices, and also fault conditions such as damaged or severed fibres.

While the concept is simple and one-button operation produces quick results, fibre testing remains fraught with complexity. Firstly, both single-mode and multi-mode fibre is in place using different wavelengths of 1310 nm, 1550 nm, and 1625 nm; and 850 nm and 1300 nm, respectively. Installations vary between local networks and long haul routes. On the one hand, your OTDR needs to view events within a few metres of the source; on the other hand, it needs enough sensitivity to trace multiple events up to 200 km away.

Source pulse width settings on a typical mini OTDR range from 10 ns to 20 ms, which in turn changes the power density of the light source and has a dramatic effect on overall performance. While OTDRs generally have no problems in identifying totally cut fibres, you’ll also need to detect more subtle fibre manufacturing or installation defects. These defects appear as kinks in the fibre (called micro- and macro-bends) that may cause only less than 1 dB losses today but can develop into hard faults tomorrow.

Even the refractive index of the fibre itself has an appreciable tolerance and this plays a large part in setting overall measurement accuracy. A 42-page pocket book from Anritsu does a good job in explaining the basics of fibre transmission and how many of these fibre and OTDR variables interact (Ref 1).

No mini OTDR attempts to handle all these variables within a single instrument. Instead, all vendors build mini OTDRs as mainframes that accept a number of plug-in modules. Each module then has optimal performance for a much narrower application. Equally, as portable products, size and weight are important and modular construction allows you to carry and power only the functions you need. Other modules can extend your OTDR’s functionality to include power measurement or visual fault location. The latter module transmits visible light that allows you to quickly observe a fibre break, especially when the break is within a few metres of the OTDR.

However, mini OTDRs don’t carry an equally “mini” price tag. Expect to pay around FFr30k/DM 9k/£3k for a mainframe and FFr 40k/DM12k/£4k for plug-ins.

OTDRs Look Remarkably Similar
The Product Survey Table overviews a selection of mini OTDRs. Vendors in general endeavour to distinguish their products from competitors’ versions, but with mini OTDRs not only do the products have broadly similar features and performance, they also look remarkably similar and produce similar displays (see Figure 1). In automatic operation, mini OTDRs display this trace together with measurement data that includes a table of each event and its characteristics. Even simpler, some models provide a pass/fail indication by automatically comparing new measurements against a template or a previously taken reference trace. Most models now also store and output data in a common format in accordance with Bellcore standard GR-196-Core (Refs 2 & 3). This feature allows you to process and archive data back at base without being restricted to using the same OTDR model.

Figure 1. A typical OTDR trace shows the general attenuation along a fibre and the reflections from various “”events’’. Eventually, either the end of the fibre or the OTDR’s inherent noise level prevents further reflections.

The standard (revised Dec 98) also provides Bellcore’s view of generic requirements that OTDRs should meet in order to satisfy the needs of service providers. The document includes benchmark test procedures that you can use to determine if an OTDR meets a vendor’s claims.

Data storage is one area where products show some divergence. Alternatives range from built-in flash cards to Gbyte hard disks, although the general approach is to treat the OTDR as a data collection unit and analyse data back at base. In any case, you need to check how many complete fibre traces one unit can store, remembering that some need to use storage for reference traces taken at installation, for example. Most vendors offer optional PC-based emulation software that enables you to process data in comfort rather than on-site. This software can also refine data collected on-site — for example, where you can collect a trace of the same fibre from opposite ends, the software can combine the two traces to provide more resolution. Or, you can combine traces on the same fibre taken at different wavelengths. You can also use the software to create templates or reference traces for taking on-site.

Selecting a Model
If you normally select an instrument by pouring over performance specifications before reaching a conclusion, you may decide against that route for an OTDR. Vendors’ mini OTDR technical specs range from a few lines to several pages. The brief versions leave too many unanswered questions, while the more complete specs require a great deal of study involving endless referral to small-print sub-notes. Anritsu, for example, provides the most complete mini OTDR specification that covers four pages and has twenty-four qualifying sub-notes. This level of spec torment exemplifies the complexity surrounding OTDR measurements and indicates why you should eye with caution the banner specs in less detailed data. Even basic OTDR specs such as dynamic range need qualification because different methods exist for expressing this parameter and one vendor even quotes two figures. As a result, selecting an OTDR can end up as more an act of faith in your favourite vendor rather than an analytical process. In these circumstances you really need to carry out a thorough hands-on product evaluation.

Despite the difficulties, it’s important to appreciate the significance of at least a few basic parameters. Dynamic range, for example, is a figure of merit that determines how far your OTDR can “look” into a fibre. “Sorting Out OTDR Dynamic Range” explains some of the background.

Beware the Dead Zone
Dead zone is another important OTDR specification and vendors’ specifications need similar close examination. In a dead zone, a reflection momentarily saturates an OTDR’s photodiode detector and the unit is unable to make measurements. The period of saturation and the detector’s speed of recovery translate directly into a length of fibre over which the OTDR can’t discern events or measure backscatter. Vendors specs quote an event dead zone to tell you how close the OTDR can discriminate between events, and an attenuation dead zone to tell you at what point the unit recovers its backscatter measurement capability.

Vendors naturally quote an OTDR’s shortest dead zones (a few metres), which occur with an OTDR set to its minimum pulse width — for example, 20 ns. But don’t forget that vendors quote their best dynamic range at an OTDR’s maximum pulse width setting (typically 20 ms), so achieving the best dead zone and dynamic range performance are mutually exclusive.

When using an OTDR it’s useful to bear in mind that, for a typical fibre with a refractive index of 1.47, a 20 ns and 20 ms pulse extend to 4 metres and 4000 metres, respectively, along the fibre. So, don’t expect to resolve close proximity events at maximum dynamic range (20 ms). And, don’t expect your OTDR to look far into a fibre at a short pulse width setting (20 ns). Additionally, the more events that a fibre shows up the shorter the distance the OTDR will be able to look into the fibre. Each reflection leaves a lower on-going pulse power to continue searching the cable. Equally, even one large reflection can also appreciably use up available forward power.

Refractive Index Sets Accuracy
Alarms bells should sound each time you see “accuracy” mentioned in mini OTDR specifications. In practice, accuracy depends more on the fibre’s refractive index, which is a figure you have to enter into an OTDR’s memory each time you make a series of readings. Because the tolerance of the refractive index greatly exceeds uncertainties in the OTDR itself, the overall absolute accuracy of all measurements broadly equates to the uncertainty of the refractive index.

OTDR’s quoted accuracy figures do provide some figure of merit though by telling how repeatable measurements will be assuming a fibre’s refractive index remains constant.

References
1. “A Users Guide to Optical Time Domain Reflectometers”, Anritsu, Luton, UK, +44-1582-433200.
2. “OTDR Measurement Records Harmonise With Bellcore Standard”, Torsten Born, Test & Measurement Europe, Aug-Sep 1997, pg 21.
3. “”Generic Requirements for OTDR-Type Equipment’’, GR-196-Core, Bellcore. www.bellcore.com.

Sorting Out OTDR Dynamic Range

Dynamic range is the ratio of minimum detectable reflected power compared to source power, and is a fundamental parameter that preponderates an OTDR’s performance. Vendors use several methods for stating power levels and, although each method is reasonable and valid, each method also produces a different result. Apart from other variables such as wavelength and pulse width that also influence dynamic range, it’s important to understand differences in the method of stating dynamic range in order to sensibly compare specs.
  At the source end, all the methods accept the level of backscattered light as a measure of source power. The main difference in the methods relates to how you define the minimum detectable reflected power. In particular, the difference depends on how you consider minimum detectable power in relation an OTDR’s inherent noise level. Several methods are in use:
• The SNR =1 method assumes the minimum detectable reflection equals the OTDR’s noise level.
• The 98% method assumes the minimum detectable reflection is 2% below the OTDR’s peak noise level.
• The Bellcore method disregards peak noise and defines dynamic range as the ability to measure a 0.5 dB splice with 0.1 dB accuracy and an event resolution of less than 1 km.
  Most vendors quote dynamic range using the SNR=1 method. The problem with this method is that the specification is more theoretical than practical. The Bellcore method aims to represent better what OTDRs achieve in practice. Vendors generally acknowledge that the SNR=1 method shows a 6 to 8 dB more favourable spec than Bellcore. You should note that the SNR=1 method assumes that an OTDR averages 3 minutes of measurements. Not all vendors’ specs mention this point.