What Happens in the First Four Hours of a Fibre Break

A cut fibre is a sequence, not a panic. Confirm the break is real, locate it from the trace, get physical access, splice, test, prove the repair. Most of the elapsed time on a typical break is not splicing — it is locating, and then waiting for permission to dig or to enter. Knowing which stage you are in tells you what to say to whoever is asking when service comes back.

Stage one: confirm it is actually a break

Before anyone is dispatched, establish that the fault is in the outside plant and not in the equipment either end. Transceivers fail, patch leads get knocked, a port gets shut by a configuration change, and every one of those looks like a fibre fault from a monitoring screen.

The quick discriminator is where the light stops. A link that shows no light at all, on every core in the same cable, on a route that has just had construction activity near it, is a physical break. A single core degraded while its neighbours are clean is usually not a cut — it is a connector, a splice going bad, or a bend that has been slowly getting worse.

This stage should take minutes, and it saves hours. Dispatching a crew to a road works site for a failed transceiver is a common and expensive mistake.

Stage two: locate the fault

This is where an OTDR earns its cost. Shot from one end, the trace shows an event at a distance along the fibre — a reflective spike and a drop to nothing at a cut, a non-reflective step at a crushed or sharply bent cable.

The critical thing that catches people out: the OTDR gives distance along the fibre, not distance across the ground. Cable is longer than the route it follows. It has slack coiled in joint boxes, it loops in and out of manholes, it takes up and down through risers, and every one of those adds fibre length that does not correspond to ground distance. A trace reading 1,340 metres does not mean 1,340 metres from the building along the pavement.

Converting the trace distance into a place to dig needs the route record: where the joints are, how much slack sits in each, and where the cable deviates. Where good route records exist, the location is usually narrowed quickly. Where they do not, the crew is walking the route looking for fresh excavation, new road markings or a contractor who has just left.

Stage three: get access — usually the longest stage

Locating the break tells you where it is. It does not give you the right to go there. Depending on where the fault falls, this stage means one of:

• A road or footpath opening, which needs the relevant permit and traffic management before a spade goes in the ground
• A manhole or duct chamber, needing confined-space procedure, gas testing and a standby person
• A private building riser or plant room, needing the building's permission and often an escort
• A live facility — transport, healthcare, industrial — where work may only be allowed in a defined window

This is the stage where repairs sit still. The crew is on site, the fault is located, the splicer is warmed up, and nothing happens because the permit is not in hand or the building manager is not answering. On a route in a public road, access can dominate the whole timeline.

It is also the stage most improved by preparation. An operator with standing permits, a known escalation contact at the facility and route records that pre-identify which authority owns each section will get to the cable substantially sooner than one starting those conversations at the moment of failure.

Stage four: prepare and splice

Once the cable is exposed the work is methodical. Strip back the sheath, identify and clean the fibres, set them in the tray, cleave, fusion splice, protect each splice and lay them into the joint enclosure with the right bend radius and slack.

The count matters here. A cut in a high-count cable is not one splice, it is every core — and each one is prepared, cleaved, spliced and protected individually. That is why fibre count drives the on-site duration far more than the difficulty of the break itself. A four-core break and a ninety-six core break are the same job repeated a very different number of times.

Two practical points decide whether the repair holds. Cleave quality determines splice loss, and a rushed cleave produces a splice that passes today and degrades later. Slack management in the joint decides whether the next crew to open that enclosure can work on it, or is fighting fibres with no length left to re-splice.

Stage five: test, and prove it

Light returning is not the end. A repair is finished when it is measured and the measurement is recorded.

The test that matters is a fresh OTDR trace across the repaired span, showing the new splice as an event with an acceptable loss, and an end-to-end loss figure for the whole link. Compare that against the pre-fault baseline if one exists. Two things can be true at once: the service is up, and the link is now worse than it was — a marginal splice will carry traffic today and fail under temperature or ageing later.

The handover record should contain the trace, the loss figures per splice, the end-to-end loss, the joint location and what was found. That record is what lets the next fault be diagnosed quickly, and it is what supports any claim against whoever caused the damage.

What actually shortens a break, and it is all decided beforehand

By the time the cable is cut, most of the repair duration is already determined. The levers are:

• Route records that are current. Joint positions and slack lengths turn a trace distance into a dig location. Without them, locating is a search.
• A pre-fault OTDR baseline. It makes the fault obvious by comparison and proves the repair afterwards.
• Spare cores on the route. Where dark cores exist and the cable is not severed entirely, service can sometimes be restored onto a spare while the permanent repair is scheduled properly rather than done in a hurry.
• Diverse routing for anything critical. The only measure that makes a single cut a non-event.
• A named escalation path. Who authorises emergency spend, who calls the building, who owns the permit — agreed in advance, not discovered at two in the morning.
• Physical protection where risk is known. Armoured cable, greater burial depth or duct protection on sections with excavation activity nearby.

The part people get wrong

The instinct during a break is to push the crew on site to go faster. On almost every job I have seen, the crew is the part already going as fast as it safely can, and the delay is upstream of them — a permit not applied for, a building contact not answering, a route record nobody can find.

What actually changes the outcome is preparation nobody sees, made months earlier: the records, the baseline traces, the spare cores, the agreed escalation. Those decisions are boring, cheap and made when nothing is wrong, which is precisely why they get deferred.

The cable that carries something you cannot lose deserves a second route. Everything else is mitigation of a risk you have chosen to keep.

— Samuel

Get the Break Fixed, and the Repair Proved

Fiber Tech Solutions is an IMDA-licensed contractor with FOA-certified splicing and termination teams, working across telecom exchanges, transport, utilities, data centres and commercial buildings in Singapore. Emergency fibre repair, fault location, splicing and certified testing, with the trace handed over as part of the job.

If a route on your network matters enough that a cut would hurt, the useful conversation is before it happens — route records, a baseline trace and an agreed escalation path.