“In fiber optics, the cable is a light pipe or waveguide, into which you inject light. If a finger presses on the pipe, it disrupts that light within the waveguide.”
And if a finger pressing on the fibre can disrupt the light signal, imagine what a backhoe can do to an optical fibre cable! As a humourous aside, you may like to read Fred Lawler’s “The 10 Most Bizarre and Annoying Causes of Fiber Cuts.”
Anyway, back to the blog. If you’re sitting in a Network Operations Centre (NOC) and you suddenly lose an intercapital link or two, you’re going to want to get to the root-cause of the problem pretty quickly. Assuming the optics are still functioning correctly and nobody is performing adds/moves/changes to patching in associated equipment rooms, there’s a fairly big chance that there has been a cable cut. But on an intercapital link, the cable could be hundreds of kilometres long, so how do you determine where to send the repair crews?
This is where spatial tools allow your OSS to provide an answer. OTDRs (Optical Time Domain Reflectometers) are able to send a light-pulse down an optical fibre to determine how far along the fibre that the cut has occurred. Let’s say that the break is at 34.6km from the patch panel that the OTDR is plugged into. This is helpful information, but two other factors will help you to identify the location more accurately.
If your OSS has a GIS that shows the exact cable route, then it is likely that it can also run a trace to estimate the GPS coordinates where the cable is broken.
You may think that if you trace the cable out to 34.6km then you’ll have the correct location. Unfortunately, you could be out by hundreds of metres because the cable / sheath length (as indicated on the GIS) and the fibre strand inside that cable (as measured by the OTDR) aren’t identical. Since the fibre strands coil around inside the cable, the fibre lengths are actually slightly longer than the cable lengths.
This difference in length is partly attributable to the helix factor. The helix factor is unique to each manufacturer / model of cable but is usually around 1-2% (see the manufacturer’s cable specs for actual values). Your OSS should have the ability to assign this factor as an attribute of each cable in your network and then consider it when tracing a route.
So assuming a helix factor of 2%, and a break at 34.6km of fibre length, then the cable sheath length will be 33.9km from the patch panel that the OTDR is measuring from. Your OSS / GIS will then show you the exact address where your repair crew can go to find the offending backhoe.
NOTE 1: Your OSS / GIS should also attempt to factor in real cable lengths, including things such as cable loops in pits (slack loops) and cable sag factors (for aerial cables) to determine the sheath length, not just geographic distances.
NOTE 2: Interestingly, larger core-count cables can actually have different helix factors depending on whether the buffer tubes are wound as the inner or outer rows inside the sheath, but I haven’t seen an OSS capable of differentiating for this yet. I guess your repair team will be able to spot the backhoe from a couple of hundred meters away from where the OSS / GIS has sent them anyway. 🙂
NOTE 3: There are a few other tricks for finding cable faults when above ground disturbance (ie backhoe) isn’t apparent, but that’s another story for another day and isn’t really in the realm of OSS anyway.
NOTE 4: Because of the various factors above that separate optical distance from sheath geo-position, it can make more sense to take the OTDR distance from the nearest even (eg splice or patch) and estimate geo from there, especially over very long spans.
NOTE 5: The Index of Refraction (IoR) impacts distance calculations too but it’s assumed the OTDR considers this before handing off the optical length to our OSS.