In short
An Optical Time Domain Reflectometer (OTDR) trace is a map of your fibre link: a sloping line for the fibre itself, with events marking splices, connectors, bends and breaks along its length. A pass means every event and the total loss sit inside the values your project specified, with results recorded for every fibre.
A fibre contractor hands over the test results, everyone nods, and the folder goes in a drawer. Months later a link fails, the folder comes out, and it turns out nobody in the building can actually read what is in it.
You do not need to be a fibre engineer to understand a test report. An Optical Time Domain Reflectometer (OTDR) sends pulses of light down the fibre and measures what reflects back, building a picture of the whole link from one end. Read properly, the trace tells you where every joint sits, how good each one is, and whether the network you paid for is the network you got.
This guide is for the commercial buyers who sign off this work: project managers and network owners in data centres, campus estates, rail and industrial environments. Home broadband faults are a matter for your internet provider, not an OTDR.
What are the events on an OTDR trace?
A trace reads from left to right, from the tester to the far end. Along the way, each feature of the link leaves its own signature.
The launch and the dead zone
The start of every trace is dominated by the test equipment itself, which is why engineers use a launch lead: a run of fibre that moves the measurement zone past the tester's own reflections so the first real connector can be seen clearly.
Fusion splices
A good fusion splice appears as a small step down in the trace with no reflection. The size of the step is the splice loss. A splice you can barely find on the trace is exactly what you want to see.
Connectors
Connectors show as a loss step with a reflective spike, because light bounces off the polished end faces. Larger spikes or higher loss at a connector usually mean dirt or damage, and cleaning is the first fix.
Bends and pinches
A loss step in mid-span where nothing should exist often turns out to be a tight bend, a crushed section or a pinched tube. Testing at two wavelengths helps confirm it: bend loss typically worsens at longer wavelengths, which is a signature engineers look for.
Breaks
A break appears as a large event followed by nothing: the trace ends early, at a distance the OTDR reads out. That distance measurement is what turns a break from a dig-everywhere problem into a dig-here problem.
The end of fibre
A healthy link ends with a clear final event at the expected distance. If the recorded length does not match the design, ask why before you sign anything: it can mean the route was not built as drawn.
What does a pass actually look like?
A pass is not a smiley face on a screen. It means every measured value sits inside the limits your project specified: the loss of each splice and connector, the total end to end loss against the link's loss budget, and the length against the design. Those limits should be agreed before testing starts, not negotiated afterwards.
Good practice, and the standard professional OTDR testing works to, is to test every fibre, not a sample, and to shoot the link from both ends. Bidirectional results matter because a splice between dissimilar fibres can look artificially good from one direction and artificially bad from the other; averaging the two gives the true figure.
Ask for the native trace files as well as the summary sheet. The summary tells you what the engineer concluded; the traces let anyone check, years later, what the link actually looked like.
Why does the report matter after handover?
The test pack is not paperwork for its own sake. It is the baseline for the whole life of the network. When a link degrades, the original trace shows what changed and where. When capacity is upgraded, the recorded losses show whether the existing fibre can carry it.
And when something fails outright, fibre fault finding starts from the handover records: an engineer comparing a fresh trace against the original can often put a fault within metres before anyone lifts a cover. Without the records, the same fault means starting from scratch on someone else's undocumented network.
If your last installation came with no test pack, that is worth fixing while everything still works: a baseline survey of a healthy network is quick, and it is the cheapest insurance a fibre estate can have.
FAQs
What is the difference between OTDR testing and insertion loss testing?
Insertion loss testing measures the total light lost end to end with a source and meter, and is the definitive pass or fail for a link's loss budget. OTDR testing maps where the loss occurs along the route. A thorough handover pack includes both, because each answers a question the other cannot.
Should fibre be tested from both ends?
Yes. A splice between fibres of slightly different characteristics can read too well from one direction and too badly from the other. Testing from both ends and averaging the results is how professional reports remove that distortion.
What should a fibre handover pack contain?
Results for every fibre, shot from both ends where specified, plus the native trace files, a splice schedule, route records and labelling that matches what is written down. If any of those are missing, ask for them before sign off.
My contractor says the link passed but I only have a summary sheet. Is that enough?
A summary is a claim; the traces are the evidence. Native trace files cost nothing extra to hand over, so there is rarely a good reason to withhold them. Ask for them as a condition of sign off.
Need fibre testing you can actually rely on?
Whether it is a new link or an undocumented estate, we will test it properly and hand you records worth keeping.