Test & Measurement - Finding fault with cables


Costly disruption and disconnection of consumers are typical consequ-ences of faults in power cables. Yet locating these faults is often difficult and time consuming. Fortunately, there are a number of test techniques available to make this task much easier. Damon Mount of Megger looks at the most useful of these

Power cable faults come in many guises. The easiest to locate by far are permanent faults on simple networks where the cable run is known, such as the supply system for a street lamp installation. That doesn't mean, however, that finding a fault is a trivial job. In fact, it can be enormously costly, especially for buried cables. Nowadays, digging a single hole in a street in a large city, for example, can cost well in excess of £100,000 and excavating a cable typically costs around £4m per mile. A better, more cost effective technique is to adopt a structured approach to diagnosing and locating cable faults, based on the use of modern test equipment.

The preliminary stage is straightforward - simply carry out continuity and low-voltage resistance checks to confirm the presence of a fault. Do not, however, succumb to the temptation of subjecting the cable to a high voltage insulation test at this stage. Doing so might alter the characteristics of the fault, and make it harder to locate with subsequent tests.

The next step is to attempt to localise the fault using a time domain reflectometer (TDR) and standard pulse echo techniques. This instrument applies a brief low voltage pulse to the cable under test and looks for voltages reflected back along the cable. Clear reflections are, in most cases, obtained from open- and short-circuit faults. By measuring the time it takes for the reflection to return to the instrument, it is possible to provide a good indication of the distance to the fault. It is always a good idea to store a reference trace before any further tests are done on the cable as any change in condition of the fault can then be seen by comparing live with recorded traces.

Dual-channel TDRs are particularly versatile, since they allow tests to be made simultaneously on two phases. The benefit of this is that a good circuit can be compared with a faulty one, which makes the results easier to interpret as joints and cable ends will also contribute their reflections to the trace. Some models, such as the Megger TDR2000/2P, can even test live circuits without the inconvenience of having to use loose, separate blocking filters

Basic TDRs are compact, inexpensive and very easy to use. They do have some limitations, but these low-cost instruments can find a high percentage of faults. They are, therefore, an excellent investment where the purchase of more sophisticated equipment cannot be justified.

It sometimes happens though, particularly in the case of high resistance faults, that the TDR cannot see the fault. Conditioning (burning) of the fault is one way to change the fault condition so that it can be seen with a TDR. This is sometimes necessary but requires another instrument, and is dependent on the cable type but can cause problems later in the fault-finding process.

A more sophisticated option is to move on to the arc-reflection method of fault location. This involves sending a high voltage pulse down the cable, which causes a temporary arc at the site of the fault. The arc is momentarily sustained by a filter built into the arc reflection test set.

Because of its low impedance, the arc looks like a short-circuit fault that can be localised with a TDR. The time interval between the high voltage pulse and the TDR pulse is critical if good results are to be obtained. For this reason, a modified arc reflection technique, known as arc reflection plus, has been pioneered by Megger.

With this technique, not one but fourteen TDR pulses are automatically sent along the cable at varying time intervals after the high-voltage pulses. The resulting TDR traces are recorded separately and, in almost every case, at least one of the traces will clearly show the distance to the fault.

An alternative way of localising faults that can't easily be seen with a TDR alone is the impulse current method. For this, the test set sends out a high-voltage pulse to establish a flashover at the fault, and the transient memory function of the test set is used to record the transients created by the flashover.

These transients travel back and forth along the cable with peaks that can be used to indicate the distance to the fault. In practice, the first reflected peak must be ignored due to the re-ionisation period, but the time interval between the second and third peaks gives a good indication of the cable length between the test set and the fault.

The techniques described so far all have one thing in common - they provide a measurement of the fault distance to the cable fault from the point of connection of the test set. Even if details of the cable run are known, this is sufficient information to determine the fault distance but not to locate the fault, as the cable rarely sits straight and horizontal in a trench or duct. In many cases accurate information about the cable run is not available. So a little further work is needed to locate the fault.

To precisely locate the fault position, a technique called pinpointing is used. This method of pinpointing faults in cables uses a surge generator - often known as a thumper in this application - to apply high voltage pulses to the cable. These pulses result in flashover at the fault location, which generates an audible noise - the thump. It also generates an electromagnetic field that can be detected by a suitable receiver.

Sometimes the thump from the fault is loud enough to be heard without any additional equipment but more commonly, especially with buried cables, a pinpointer is used. This is basically a sensitive ground microphone connected to an amplifier and headphones. The user simply moves the pinpointer along the cable run until the thumping is most clearly heard and the magnetic field is strongest. This should be the fault location. 

Faults with cables in ducts can be difficult to find, however, as the sound can travel down the duct making the listener less able to pinpoint the exact fault location. It is at least easier and less costly to replace a section of cable in a duct than dig up a direct buried cable. Although many faults in power cables are high resistance faults where the thumping technique is very useful, it's worth mentioning that not all cable faults will thump. Short-circuit faults for example do not flashover, so no electromagnetic field is formed and because the energy of the pulse is not dissipated in the form of sound, there is no thump to locate.
In this instance a TDR and a cable route tracer can be used to find the distance to fault, but locating the exact site of the fault is more difficult. This is why the low voltage tests are applied first, before conditioning causes a resistive fault that may flashover to become a short circuit that won't thump.

No one would claim that locating faults on power cables is easy but there are many types of test instruments now available that, when used in conjunction with a structured approach to fault location, will provide assistance in locating even the most intransigent of faults. The days of the black art of cable fault locating are past, because it is now too costly and too time consuming to go down this route. Since the faults themselves often lead to downtime and the associated consequential losses, money invested in the latest cable fault location equipment is money very well spent indeed!

Data Centre news from our sister title