Choose range pulse width and averaging settings

What you should be able to do

Explain the different purposes of OTDR range, pulse width and averaging. Choose a reasoned starting configuration under the actual test procedure, then verify that the acquired trace supports the required measurements.

Teaching

Range, pulse width and averaging interact, but they do different jobs. Range defines the acquisition distance window. Pulse width influences pulse energy and the ability to distinguish nearby features. Averaging combines acquisitions to reduce random noise. A useful setting must suit the actual path and measurement objective; no single preset covers every campus, industrial or short premises link.

For a complete-path acquisition, account for the launch fiber, installed link and receive fiber. Select an available range that captures the required path and permits its far end to be identified under the manufacturer's procedure. A large range setting alone does not provide enough optical signal to reach the end. Confirm the trace instead of assuming that a displayed distance scale proves coverage.

A shorter pulse can preserve more detail around closely spaced events, but carries less energy at comparable pulse power. A longer pulse can improve the returned signal and usable reach while increasing dead zones and merging nearby features. Use the shortest pulse that supports the required measurements with adequate signal under the procedure. Some work requires complementary acquisitions or an approved automatic multi-pulse method.

Longer averaging can improve signal-to-noise performance. It takes more time and cannot repair contamination, correct an erroneous group index, extend a truncated acquisition window, or recover spatial detail hidden by an overly wide pulse. Do not confuse a smoother trace with a fully resolved trace.

Start from the documented instrument procedure or approved automatic mode, confirm the settings actually used, and inspect the result. Look for the expected endpoints, useful backscatter around events, excessive noise and unresolved regions. Change the relevant setting deliberately and retain trace identity and settings. Do not weaken acceptance limits or suppress event reporting merely to make a display look clean.

Worked through

Launch fiber: 0.2 km. Installed link: 4.0 km. Receive fiber: 0.2 km. Total path to the end of the receive fiber: 0.2 + 4.0 + 0.2 = 4.4 km.

A 4.0 km acquisition range cannot capture that whole path. Choose a suitable available range under the instrument instructions and verify the actual endpoint. The exercise does not prescribe a particular menu choice, fixed range multiplier, pulse duration or averaging time.

The two 0.2 km accessory lengths are exercise inputs. They are not universal launch/receive recommendations. Their suitability also depends on the measurement conditions, including pulse width and dead-zone behavior. Lesson 292 addresses launch and receive fibers in more detail.

Reading The Diagram

The top strip shows the three consecutive path segments; its widths are not proportional to the fictional lengths. The pulse drawings show duration differences, not real instrument waveforms. The averaging sketches are qualitative views of noisy and smoother declining signals, not measured traces or acceptance evidence. No event loss or pass/fail threshold can be read from them.

Original Practice Cases

1. The far end is cut off at the right edge. Check the acquisition range and expected total path. Extending averaging alone cannot put an out-of-range endpoint inside the window.
2. The whole path fits, but the far-end backscatter is noisy. Inspect the setup and consider appropriate averaging and pulse adjustments under the procedure. Preserve the original acquisition for comparison.
3. Two nearby connections merge after a longer pulse is selected. The extra reach has reduced usable detail. Obtain an appropriate additional acquisition or approved analysis rather than declaring that one connection disappeared.
4. A trace is smooth but the group index is wrong. Smoother data does not correct the distance conversion.
5. The same long-pulse preset is copied from an outside-plant route to a short rack link. Reassess resolution and accessory requirements; the former setting is not automatically suitable.
6. A lower event-reporting threshold creates many questionable events in a noisy region. Investigate trace quality and settings. A detection threshold is not the same thing as a contractual acceptance limit, and setting it does not guarantee reliable detection.

Supervised Comparison

Use an existing training dataset or isolated supervised training setup. Hold fiber identity, direction and wavelength constant while comparing the intended setting change. Record what became clearer and what became less distinguishable. Retain all traces required by the plan. Follow optical-safety and connection procedures; do not connect to a live service without the specifically authorized equipment and method.

Record Sheet

For each acquisition, capture fiber ID, direction, wavelength, group-index setting, range, pulse width, averaging or acquisition time, launch/receive arrangement, instrument identity and time. Note whether the far end is visible and whether each required event can be evaluated. Explain why another acquisition is needed when one trace cannot answer all questions.

Knowledge Check

Where beginners go wrong

Sources