The trace is a plot of what comes back. A pulse is sent along the cable and the reflections returning from impedance changes are recorded against time. Because the pulse travels at a known speed for a given cable construction, time can be converted into distance, and the trace becomes a picture of the electrical route from the test end to the far end and beyond.

How a reflectogram is formed

Every change in the characteristic impedance of the cable reflects part of the pulse. The characteristic impedance depends on the geometry and the materials, so any point where those change produces a reflection. A joint with slightly different dimensions, a repair using a different cable type, a termination, and the far end of the cable all qualify, along with any fault that alters the local impedance.

The amplitude of the reflection is proportional to the size of the impedance change, and its polarity depends on the direction of the change. A fault that reduces the impedance, such as a short between conductors, reflects with one polarity. A fault that increases it, such as an open circuit, reflects with the opposite polarity. The far end reflection has its own characteristic polarity determined by how the cable is terminated.

What the instrument displays is therefore a series of events along a distance axis. The skill in reading it is in attributing each event, and attribution depends on knowing the route rather than on the trace alone.

Velocity of propagation and the settings it changes

The velocity of propagation is the speed at which the pulse travels, expressed as a proportion of the speed of light or as a distance per unit time. It is determined by the insulation material and the cable construction, and it is the constant that converts the measured time into a distance.

An incorrect velocity produces a trace whose shape is correct and whose scale is wrong. Every event appears at the wrong distance, and the proportion of the error is the same throughout the trace. That behaviour makes the error detectable: if the far end of the trace does not coincide with the known route length, the velocity is wrong rather than the cable.

Where the velocity is not known from the cable data, it can be established by testing a known length of cable of the same type, or by adjusting the setting until the far end reflection coincides with the recorded route length. Both approaches assume the route length is known, which is the second constant the trace depends on.

Reading reflections and their polarity

Reading a reflection means establishing three things about it: its distance, its amplitude and its polarity. The distance locates the event, the amplitude indicates the size of the impedance change, and the polarity indicates the direction of the change.

A large reflection with the polarity associated with a low impedance fault indicates a short between conductors or between a conductor and the sheath. A large reflection with the opposite polarity indicates an open circuit or a break. A small reflection with either polarity is more likely to be a joint, a repair, a change of cable type or a moisture condition.

The amplitude comparison between two events on the same trace is more reliable than the absolute amplitude, because the absolute value depends on the pulse energy and the cable attenuation. Where two reflections appear similar in amplitude and opposite in polarity, the pair often corresponds to the two ends of a section of cable that has different construction or that has been repaired.

Distinguishing a joint from a fault

The distinction is made from the route record rather than from the trace. A reflectogram with a small reflection at 340 metres is a trace with an event; whether that event is a joint depends on whether the route has a joint at that position.

Where the joint schedule is available, the trace can be read against it. Joints that appear exactly at their recorded positions confirm the velocity setting and the route length, which in turn gives confidence in the position of any unexplained event. Joints that appear at positions that do not match the schedule indicate that either the schedule or the velocity is wrong, and the discrepancy has to be resolved before any fault distance is relied on.

Where no schedule exists, the interpretation becomes probabilistic. A small reflection consistent with a joint construction and consistent with the spacing of other joints is likely to be a joint. A larger reflection, an event that appears close to the near end, or an event whose position is inconsistent with the route features is more likely to be a fault. The distinction is a hypothesis to be tested by pinpointing, not a conclusion.

Interference and how to recognise it

Interference appears on a trace as reflections that do not correspond to anything on the route. The commonest sources are nearby services that couple to the test arrangement, parallel cables carrying load current that induce voltages in the cable under test, and, where a surge is being used, discharges occurring at poor connections in the test lead rather than at the fault.

Interference has signatures. Events that appear at the same distance on every trace regardless of changes to the cable, events that shift when the test lead is moved, and events that appear below the near end of the cable or beyond the far end are all indicators. Repetition of the measurement with a different lead arrangement is the simplest way to identify them, because genuine route features stay in place while coupling artefacts move.

Where interference is persistent, the record should state it, because a later trace will contain the same artefact and the comparison has to account for it. A trace marked with the known interference positions is more useful than a clean-looking trace whose unexplained events are re-investigated each time.

Pre-location accuracy and its limits

Pre-location accuracy depends on three things: the velocity setting, the accuracy of the known route length, and the resolution with which the reflection can be identified. All three degrade with distance, because the pulse attenuates and disperses as it travels and the reflected events become both smaller and broader.

The practical expression of accuracy is a proportion of the route length. A one percent accuracy on a two kilometre distribution circuit is twenty metres, which may be within a single excavation. The same proportion on a twenty kilometre transmission circuit is two hundred metres, which is not a position at all.

That asymmetry is why pinpointing exists. Pre-location narrows the search to a section, and pinpointing identifies the position within that section. The record should state both, because the pre-location result is what justifies the search area and the pinpointed result is what justifies the excavation.

Combining with pinpointing equipment

Pre-location and pinpointing equipment are normally supplied as a set, because the method has to change with the fault condition. A trace taken with a low-voltage pulse identifies a low-resistance fault. A high-resistance fault requires the surge source to break the fault down and provide a reflection, and the same surge source then drives the acoustic or electromagnetic detection used for pinpointing.

Where the two are combined in one arrangement, the transition between them should be recorded. The trace used for pre-location, the surge settings used, and the pinpointing method and result together form the record of the location, and a record missing any of the three cannot be repeated.

Recording traces for later comparison is the last discipline and the most neglected. Storing the trace as raw data rather than only as a distance in a logbook allows it to be re-read when the route changes, when a joint is added or when the velocity assumption is revised. The cable construction details that determine the velocity are documented by the manufacturers, such as the material published by Prysmian and Nexans, and the cable classes themselves are defined in IEC 60502-2 for medium voltage extruded cables and in IEC 60840 for the higher voltage classes. Field experience with cable diagnosis is coordinated through CIGRE study committees.

Recording traces for later comparison

The record should identify the circuit and the test end, state the velocity setting and the route length used for scaling, and store the traces from both ends where both were taken. Testing from both ends is a check that costs little and resolves the ambiguity caused by attenuation, because an event that is close to one end is distant from the other.

Where a trace is taken as a baseline before any fault occurs, it should be stored with the same detail. A baseline trace taken at commissioning identifies the joints and the terminations, and it makes every later trace interpretable. Without it, each investigation starts by establishing what the route’s normal reflections look like.

The full equipment range for pre-location and pinpointing is grouped on the cable fault testing hub. The wider procedure that these traces feed into is covered in the accompanying article on how fault location proceeds from pre-location to pinpointing.