Single-Ended and Double-Ended Localization: Accuracy Limits of Fault Location Devices for Small-Current Grounded Systems

Nov 21, 2026

Leave a message

Here's the thing about accuracy specs for fault locators: one number on a datasheet won't tell you much about how the device will actually perform on a real distribution line. Single-ended and double-ended locating work on completely different principles, and each one runs into its own kind of error. If you're comparing equipment, it pays to know where each method stops being reliable instead of chasing one isolated figure.

Single-ended locating measures from one end of the line only. A fault sends a traveling wave down the line. The wave hits the fault point, bounces back, and the device uses the time gap between the first wavefront and the reflected one, along with wave speed, to figure out the distance. The math is simple enough. The hard part is telling the reflected wavefront apart from everything else. In a small-current grounding system, the reflection from the fault point is often weak. Distribution lines also have branch points, line ends, and impedance discontinuities, and every one of those creates a reflection. Several wavefronts can overlap on the time axis or get buried in noise. The device may see a signal and still have no reliable way to know which wavefront came from the fault. When the fault reflection lines up with a reflection from the remote bus or from a healthy line, single-ended locating hits a dead zone. Wave speed adds another error source that's easy to overlook. It changes with line type, sag, and soil conditions, so one fixed speed setting will create a systematic offset across different sections. On simple overhead lines, single-ended locating can get down to the few-hundred-meter range. Put it on a distribution line with many branches, and identifying reflections gets much harder, so accuracy and reliability both drop. Its limit is basically set by how complex the grid topology is.

Long-Distance Overhead Lines in Rural Distribution Networks: Challenges in Deploying Small-Current Grounding Fault Location Devices
Can a low-current ground fault location device be used in underground coal mine power supply systems?

Double-ended locating avoids the reflection-identification problem. Devices at both ends record when the first fault-generated traveling wave arrives. The system uses the time difference and the total line length to calculate the fault position. It uses only the first wavefront and doesn't depend on later reflections, so in theory there's almost no blind zone, and the accuracy ceiling is higher than with single-ended locating. But that ceiling comes with a hard condition: both ends have to be on the same clock. Double-ended traveling-wave locating usually needs time synchronization within 1 microsecond. High-precision cases may need to get close to the 10-nanosecond level before the location error drops to meters. Any sync error beyond that window gets amplified by wave speed. A 1-microsecond sync error translates to roughly 150 meters of spatial error at traveling-wave speeds near the speed of light. GPS loss, poor satellite visibility, or internal clock drift can each push synchronization outside the allowed range. Once that happens, the accuracy promise of double-ended locating loses its foundation. So its limit isn't really in the algorithm. It's in whether the synchronization link can be trusted all the time.

At the selection stage, these two limits point to different judgments. If the line doesn't have many branches and the remote end has reliable communication and synchronization, a double-ended setup can push accuracy to a level single-ended locating struggles to reach. But real distribution grids often don't line up that way. Monitoring points aren't always sitting at both ends of the line, communication conditions vary, and putting double-ended time-sync equipment on every possible fault section is neither practical nor economical. A single-ended device can work on its own without needing a partner at the other end, so it remains a workable choice on lines with complex branching and limited communication. You just need to be honest about what accuracy to expect: it gives you a usable fault range, not an exact point. Knowing that limit helps more on site than chasing one isolated high-precision number.

Send Inquiry