The age profile of low-voltage cable within the Northern Ireland Electricity network ranges from the late 1930s to present day, and the cable type consists of a mixture of paper-insulated lead-sheathed cables and various forms of polyvinyl chloride-insulated cable. Some of the utility's oldest assets are located in provincial towns and cities where recent development work has provided an opportunity for urban renewal. In the event of a fault or emergency situation, the ongoing priority of Northern Ireland Electricity (NIE) — the distribution network operator for Northern Ireland — is to restore its customers' supply within a short time frame, safely and within financial and regulatory limits.

It also is now paramount for the utility to satisfy the legislative requirements associated with excavation and reinstatement, to minimize disruption to the public, traffic and general infrastructure. During the past five years, in many towns and cities within Northern Ireland, decorative stone-surface footways have been laid on top of concrete. These areas have highlighted the need to reduce the excavation and reinstatement required; locating faults in these areas must be accurate to minimize the impact of multiple excavations.

NIE experiences an average of 300 faults each year on its low-voltage mains and service underground cable system. These are faults where time-domain reflection (TDR)-type equipment traditionally has been used to provide a location prior to excavation work. Whether these cables form part of an interconnected circuit or not, it is essential prompt action be taken to locate and repair the fault to ensure system security. In addition to these faults, an average of 250 transient-type faults occur each year. The transient-type faults are characterized by a ruptured fuse causing a loss of supply, and, in each case, further investigation is required to establish the cause. This investigation work also is usually completed using TDR-type equipment.

Cable Fault Location

For the past 50 years, TDR technology has been available to locate the position of underground cable faults. Improvements have been made to the technology, and TDR equipment is now smaller and portable. Even though features such as additional memory and the ability to compare cable traces have been added, fault location teams still experience essentially the same problems with this equipment as their predecessors.

TDR diagnosis involves the collection of often extremely complex waveforms that represent the reflections from the cable circuit being tested. This data is then analyzed in an attempt to distinguish reflections attributable to the cable's features from those attributable to the fault position. Following the on-site decision-making process, a cursor is used to identify the fault reflection time and distance to the fault.

In common with all U.K. distribution network operators, NIE is faced with ongoing cost constraints. With multi-skilled staff no longer dedicated full time to cable fault location, the need for accurate fault location equipment has increased. Distribution network operators also face significantly greater regulatory and customer pressures to locate and repair faults in minimum time.

Traditional TDR equipment presents problems to those tasked with finding faults on the low-voltage distribution network. Much of the TDR equipment available was designed to test telecommunications cables with minimal modifications to support their use on power cables, while others were designed for exclusive use on medium- and high-voltage cable applications. For these reasons, NIE was keen to test Kelvatek's Reflekt product, to assess its ease of use and ability to automatically locate faults on complex low-voltage networks.

Cable Fault Locator

Drawing on 30 years of design experience, Kelvatek developed the Reflekt cable fault locator — based on TDR principles and techniques originally developed at the University of Strathclyde — following almost 10 years of research. Kelvatek manufactures and supplies a range of products incorporating cutting-edge technologies in the power and rail industries to solve many key problems:

  • Underground cable fault location

  • Distribution fault management and conditioning

  • Circuit breaker analysis

  • Safety critical equipment

  • Auto-reclosing systems for distribution and rail signal networks.

With a focus on the automation of test processes, this product range is designed to minimise the potential for user error, reduce site testing time and ensure the equipment can be used by less-experienced staff. Rather than requiring the user to diagnose, compare and interpret a series of traditional TDR waveforms, Reflekt's patented algorithms automate this process. The user is provided with a simple distance to fault, a positive advantage for low-voltage network applications where the mix of single- and three-phase joints can confuse even the most experienced fault location engineers.

Evaluation Process

NIE decided to trial the Reflekt auto TDR using some of its more experienced fault location staff to undertake a comparison of the unit and its success rate with NIE's existing TDR fault location products. Two experienced staff responsible for working in areas that included rural villages, small towns and city center networks were selected to undertake the trials. Over several months, both engineers recorded the fault locations found by using their traditional TDR instruments and those found using the new Kelvatek product.

TDR traces obtained from the low-voltage distribution network are extremely complex, and there are many reflections to analyze that may or may not be the fault. The earliest results from the trial revealed auto TDR should be used only after following completion of the fault location process using the manual TDR, to avoid unduly influencing the user as to which waveform feature to select.

The evaluation process then became a case of testing the fault location equipment on as many faults as possible — both as they occurred in the normal course of events, but also wherever possible by visiting areas with difficult fault conditions to allow a large number of test comparisons to be made quickly.

As is generally the case with TDR devices, the evaluation process used this fault locator primarily as a pre-location tool to indicate the fault position within a few meters, allowing an assortment of experience, pinpointing tools and techniques to be used for fine location. This eliminates the usual errors caused by cable routes and variances in cable records.

Field Testing Results

During the early evaluation process, the Reflekt was tested on more than 30 cable faults. For faults on a three-phase cable, it achieved a 100% success rate in correctly locating the fault position, exceeding both the anticipated success rate of an experienced engineer using a manual TDR and Kelvatek's estimated success rate of 80% from a single test.

It is often necessary to conduct tests on a single-phase service cable (for example, when testing from a residential customer's premises), a situation that generally presents the most difficult TDR waveform data to interpret. NIE was able to test the location equipment in this situation on more than 10 occasions, and it either matched the older TDRs in accuracy — or was more accurate in locating the fault position. Two reports on NIE applications are presented to illustrate the superior performance of the Kelvatek equipment.

In the first example, one of NIE's residential customers in Antrim reported a loss of power to their home and no other adjacent customers were without a supply. A visit to the source substation confirmed no fuse failures, so NIE staff engineers then visited the customer's house to verify the initial assessment. Tests were conducted from the service cut-out to the mains cable joint. The conventional manual TDR indicated a distance to fault of 2.5 m (8.2 ft), but use of the Reflekt TDR gave a fault distance of 11 m (36.1 ft). This second distance corresponded closely with the expected location of the mains-service T-joint and subsequent excavation revealed a faulty joint. Cable faults close to the source are always difficult to locate accurately on a manual TDR, but this example illustrates the benefit of using the Kelvatek TDR.

The second example occurred in the town of Ballymena where a cable fault disrupted supplies to many customers, although some customers supplied by the same cable remained on supply. Tests were conducted from a pole-mounted distribution point, but, as the cable network had multiple branches and many single-phase service connections, a diagnosis was extremely difficult. The manual TDR gave a distance to fault of 138 m (453 ft) and the Reflekt TDR a distance to fault of 231 m (758 ft). Physical examination of both distances showed recent ground disturbance at around 230 m (755 ft) from the test point, so the fault team decided to excavate at the predicted position. No fault was found, but further tests on the cable showed it was dead at this point, indicating the cable had been severed so a further test back toward the substation had to be conducted. In this case, the results of these tests using the conventional TDR and the Reflekt TDR test were identical at 9 m (30 ft) confirming the location of the cable fault.

In this instance, the incorrect reading at 138 m was probably the end of one of the cable branches from the mains cable. Although the location tests in this example required two excavations to locate the fault using the Reflekt TDR, it may well have required multiple excavations using the results from the conventional TDR.

With virtually all power cable faults, some error in the predicted fault position can be expected. Cable records can span a period of more than 80 years, and cable routes are not always clearly defined, or new cable sections have been inserted without the records being updated. To compound this problem, cables often can show varying characteristics, especially the velocity at which a TDR signal will travel along the cable. The usual practice is to use TDR as a pre-location method, and then use another technique such as thumping with an auto recloser to pinpoint the fault. Unfortunately, in the two examples described, open-circuit faults existed that could not be pinpointed easily using such techniques, so the accuracy of TDR proved valuable in locating the fault location position.

Successful Fault Location

NIE's field experience used to evaluate the Reflekt unit can be summarized as follows:

  • Improvements in accuracy of pre-location of fault positions and the elimination of some of the complexities of the low-voltage distribution networks

  • A definite indication of the fault location within 3 minutes

  • Reduction in the uncertainty of the fault position

  • Fewer excavations to locate and repair faults.

The reduced excavation results is improved network reliability, with customers having supplies restored more quickly and with less disruption to the local economy and environment. NIE benefits from a reduction in the fault repair costs.

Throughout the field trials, Kelvatek's TDR equipment was used by, and compared with, the performance of skilled, experienced fault location engineers. However, they used the location equipment in full automatic mode, whereby only simple information about the circuit being tested was supplied before the test equipment was able to collect, analyze data and report the fault distance. This process now can be undertaken by less-experienced staff, allowing first-response teams to locate faults with a high degree of success without the need for experienced staff, who can then remain available to locate the more complex faults on the network. The de-skilling of the cable fault location activity will further reduce NIE's network operational expenditure.


Kevin McDowell (kevin.mcdowell@nie.co.uk) is the fault and emergency manager for Northern Ireland Electricity.

Companies mentioned:

Kelvatek | www.kelvatek.com

Northern Ireland Electricity | www.nie.co.uk