By Walter Wittig
Cable Fault Locating - like a challenging, strategic sport, requires a good working knowledge of the game's basics and an understanding of the opponent's shortcomings -as well as your own team's. Here's a tip to help play the game.
Why Didn't My TDR Find the Power Cable Fault?
The TDR (Time Domain Reflectometer), also known as a cable radar, echometer, or scope, is an extremely helpful device for cable trouble analysis, if used within it is limits of capability. Very simply, a TDR transmits a small pulse out onto a cable. The pulse travels to the end of the cable and then gets reflected back to the set from say, the open end, returning an upward echo or signature, to represent the open end. This is displayed on a screen where the vertical direction represents resistance (impedance) and the horizontal direction represents cable length. If there is an irregularity in the cable's makeup - say at a splice, fault, etc., then there will be a small reflection from it also, back to the set. However, the most easily recognised irregularities that it displays are opens and shorts (Fig. 1).
Another seldom understood limitation is that the TDR is best at finding severe opens and low resistance shorts only within the capability of it's output voltage. A TDR typically puts out a very narrow pulse (millionths to thousandths of a second) at 10 or 20 volts maximum, depending on range (highest range has highest voltage pulse). Faults that show up as "0" on a Megger at 1kV, but 1000 to infinite Megohms on a 9v Ohmmeter, will not likely show up on the TDR. In fact, a fault of over 1000 Ohms at low voltage is just about the lowest resistance that can be seen easily by a TDR operator. It's not that a TDR cannot "see" or display the fault, but the operator cannot easily distinguish the fault from the other irregularities on the cable, which invariably show up (Fig. 2).
The reason that the fault is sometimes hard to see relates to the ratio between fault resistance and the cable impedance. Power cable has a characteristic impedance of roughly 30 - 60 Ohms at the TDR's narrow pulse frequency. Ten times the cable impedance becomes the approximate limit at which you can barely see a reflection on the TDR. For power cable, this means that, depending on range, you cannot easily see low voltage faults of 300 - 600 Ohms across the cable being tested. At 1,000 Ohms, you have very little chance of success.
Similarly for Cable TV coaxial cable of 72 Ohms, the limit would be around 720 Ohms. As you approach these higher fault resistances, the experience level of the operator comes into play. Someone with a lot of experience at interpreting small irregularities (higher resistances) will have more success than someone who looks at a TDR once a year.
Another problem with cable irregularities relates to the type of cable being viewed. Coaxial cables and their relatives in the power industry - concentric neutral cables - give the best TDR displays (Fig. 3A) because they are manufactured to tightly controlled specifications for insulation thickness between the centre conductor and shield. This is because coax needs to carry high frequency signals with low loss, and high voltage concentric cable needs to transmit high voltages with low stress and no chance of breakdown. Their images on the TDR appear closest to a straight line, except where faulted, spliced, through (switched) connections are made and transformers are situated.
Secondary cables, on the other hand are often thrown haphazardly into the ground, with widely varying distances between conductors, resulting in an equally wavy trace on the TDR. If you then add a high resistance fault to the wavy trace (or sometimes even one fairly low in resistance), it sometimes results in not "seeing" the fault among all the other reflections (Fig. 3B).
Most faults above a few hundred Ohms on secondary cable are difficult to detect without other techniques being employed. For instance, if the TDR has phase or good/bad trace comparison, the chances of picking out a bad spot are increased, as they are if you can carefully burn a stubborn conductor under controlled conditions, to create a carbon bridge which would look like a lower resistance than before. For this, trace storage can be very useful.
Other fault, if not detectable, are often only located in conjunction with more advanced high voltage and other test methods or practices.
Most cable crews who don't have much test equipment, determine that a cable is faulted by various isolation techniques, sometimes dividing the circuit up until only the underground section in question blows the fuse or breaker - a form of reclosing.
Reapplication of powerful 60Hz system energy (as calculated in fault current analysis) is capable of creating further serious damage to a cable that originally might only have had a small pinhole fault in it. To make matters worse, cables running right next to the reclosed one might also suffer unintentional damage from the virtually unlimited system energy supplied before the fuse or breaker reacts.
A much less disruptive way of checking which section is faulted, is by using a MeggerȘ unit. Many faults on power cables exhibit very high resistance (infinity to many tens of Megohms) with a normal low voltage Ohmmeter using 1.5 to 9 volt batteries as their current source. This is therefore useless for determining which cable is faulted. However, with the Megger on an appropriate voltage range, the bad cable is usually identified quickly.
If a "0" Megger reading is used as the basis to determine if a cable fault can be seen on the TDR, the maintenance crew will likely be very disappointed. Most power cables do not burn badly enough at the fault to "weld" together as the Megger would lead you to believe. Current protection usually cuts in fast enough to minimize this sort of damage. If a second check is made with a low voltage Ohmmeter, there is a chance that the fault will show up well on the TDR if it is well below 1000 Ohms - the lower, the better.
It is not within the scope of this article to give a lengthy explanation of another technique now used to quickly locate high resistance faults. This involves the ARC Reflection Method (ARM) which allows a TDR to actively view the fault while a capacitive discharge set (thumper) causes a momentary fault flashover, making it easy to identify the fault and distance.
Further information can be supplied on request.
Walter Wittig has been involved with cable testing and fault locating for the past 19 years. He currently heads Cable 3000, specializing in training, consulting in cable testing and sale of test equipment. He can be reached at (905) 427-3000. ET