Today testing fiber optic networks is more important than ever. As bandwidth speeds increase, the testing thresholds become much more stringent compared to lower speed networks. One test that we would like to highlight for discussion today is Optical Return Loss. Optical Return Loss or ORL is one test that should not be overlooked due to the tremendous impacts it can have on a Singlemode system.

When we talk to customers about ORL many folks have a misconception of what ORL is and the impact of the measurement. ORL is defined as a measurement of the power of the injected light compared to the power of the light reflected back to the source. An ORL measurement will incorporate all elements of the fiber link including connectors, mechanical splices, in-line devices and the Rayleigh scattering that is intrinsic to fiber. Note that an ORL measurement should always be a positive value, further away from zero indicates better performance. When testing 10 Gbps networks a threshold of +27dB is required, although installers typically strive for +30dB measurements.

Network ORL Minimum ThresholdsLinks which do not pass an ORL test will typically experience transmission problems. ORL can affect (a system) in several ways including: low signal to noise ratios, laser temperature instability, laser output fluctuations, receiver interference, which all lead to high bit error rates.   Since ORL can have such a significant impact on the network, it is important to make sure that the link is properly tested.

Testing for ORL can be performed by two different types of measurements. A true ORL tester such as the Exfo Max Tester or Kingfisher 7340 can provide quick and accurate measurements while testing for insertion loss simultaneously. A link can also be measured for ORL by using an OTDR. But note that the OTDR measurement has typically been seen as +/- 2dB; this test is more of an estimation. In most situations the ORL meter will be the preferred measurement style, although exceptions do apply.

Cleaning and inspecting fiber optic connectors has one of the greatest impacts on a system’s ORL measurement. When light changes speed between glass and air a reflection, or in fiber’s case, a back reflection is generated. This back reflection is generated at connectors, mechanical splices or other events where light changes speed. Fusion splices are a non-reflective event which makes them ideal for improving ORL measurement.   Note that dirty or damaged connectors can greatly reduce ORL measurements as the connector performs poorly, it generates high back reflections.

If a span of fiber fails an ORL measurement in either direction it is important to identify the event or element within the system which is generating strong back reflections. The proper tool for this troubleshooting project is an OTDR. By using an OTDR, a skilled technician is able to measure and classify each event within the fiber span. A measurement known as Reflectance allows the technician to analyze this data and identify contaminated or damaged connectors, breaks or bad mechanical splices which can cause poor ORL. Reflectance measurements will indicate the amount of back reflection a reflective event creates. Note that Reflectance measurements are negative instead positive as ORL. Numbers further away from zero in the negative spectrum indicate better performance.   A UPC connector should display a reflectance level between -50 to -55dB in most cases. There are of course exceptions such as long haul testing, where the connector may only register at -40dB of reflectance. Skilled technicians can identify poorly operating connectors with ease using an OTDR.   Reflectance values are listed in the image below.

Reflectance value mated connectors

Keeping ORL Simple in the Field

 

A few months ago we received a phone call from a customer with a couple questions about ORL which were a bit out of the ordinary. Our customer, an experienced fiber installer and project management group, was dealing with a span which failed ORL in both directions.   OTDR test results indicated the one reflective event approximately 40km into the span did not meet Reflectance value thresholds in either direction. The end user stated that based on some specific formula they use, the poorly performing connector should have only been affecting the measurement in one direction since the connector was so far from one end of the span. After reviewing several white papers and manufacturer resources we were unable to locate or even find a mention of this formula the end user insisted was valid.   Weeks later at a trade show, we listened to a presenter discussing this very topic.  Their argument for validating the acceptance of poor reflectance values based on the event location to the transmitter was less than reassuring.

We never heard back from customer to see how this issue got resolved, but we can share some good advice to remove you from these types of situations, both as the contractor and the end user. Look at ORL as a total system measurement which incorporates everything in the system. If a fiber span fails ORL, something in the span isn’t functioning properly.  Test the span with an OTDR. If a reflective event is found which does not meet the industry reflectance thresholds, fix that event! The general consensus among installers, test equipment manufactures and testing specialists is this, “If all elements of a span,  including fiber and events meet the individual reflectance thresholds, then the span should pass ORL.”   Again exceptions due apply, but those are few and far between.

 

Conclusion

An ORL measurement can be a great metric to use to get a preliminary idea of how the system will handle high speeds. Although other factors like Chromatic Dispersion and Polarization Mode Dispersion have significant impacts on performance, ORL is one factor that can be influenced by the installer through a quality installation. Dirty or damaged connectors and bad mechanical splices will be the most common issues an installer faces when dealing with poor ORL values. Make things easy on yourself by always cleaning and inspecting connectors before ORL tests. If mechanical splices are an issue in the system, replace that mechanical splice with a fusion splice and watch that ORL measurement improve.