Field Testing of Medium-Voltage Cables
Cable Degradation and Diagnostic Tests
This overview is provided as an insight into understanding of the nature of in-service cable degradation and some of the more commonly used diagnostic techniques commercially available for performing tests in the field on extruded, medium-voltage, shielded power cables. The objective of any diag- nostic test is to identify, in a nondestructive way, a potential problem that may exist with a cable, so that preventative action can be taken to avoid a possible in-service failure of that cable. This assessment applies to cable sys- tems comprising of the cable itself and the associated accessories such as splices and terminations. Field diagnostic tests can be performed on cables during various stages of their existence. The IEEE std 400-2001 defines these tests as follows:
Installation test: Conducted after the cable is installed but before any acces- sories (joints/splices and terminations) are installed. These tests are intended to detect any manufacturing, transport, and installation damage that may have occurred to the cable.
Acceptance test: Performed after the installation of all cable and accessories, but before energizing the cable with system voltage. Its purpose is to detect installation damage in both the cable and cable accessories.
Maintenance test: Also referred to as after-laying tests that are performed during the operating life of the cable system. Its purpose is to assess the con- dition and check the serviceability of the cable system so that suitable main- tenance procedures can be initiated.
The IEEE std 400-2001 also defines cable field tests into two main groups, Type 1 and Type 2 tests.
Type 1 field tests: These tests are normally performed at elevated voltages and are a pass/fail type test. The traditional high-potential (hi-pot) test is an example of a Type 1 field test. The cable either passes or fails the test, but one establishes little knowledge of the condition of the cable other than whether the cable system withstood the voltage for the duration of the test or not. This test is beneficial in that it is normally able to root out severe defects in a cable. Many defects, however, may pass undetected during a pure voltage-with- stand test.
Type 2 field tests: These cable diagnostic tests are performed at test voltages above and/or below the normal operating voltage of the cable. These tests assess the condition of the cable system and try to establish the remaining service life. There are two categories of Type 2 cable diagnostic tests available:
(1) Tests that assess the overall (integral) condition of the cable; and (2) tests that detect and locate discrete defect locations in a cable system.
In recent years, a great deal of research and development has focused on field cable diagnostic tests. This effort was due in part to the fact that many of the new PE and XLPE cable systems installed in the late 1960s, 1970s, and early 1980s were prematurely failing as compared to the PILC predecessors. Traditional DC hi-pot testing was not only found to be ineffective in trying to diagnose the failure mechanism before cable failure occurred, but the presence of these elevated DC test voltages was also found to be potentially damaging to PE and XLPE service-aged cables. Whereas many PILC cables were lasting well over 50 years before being replaced, some of the originally installed PE/XLPE cables were experiencing failures within 10–12 years of their service life. A concerted effort to determine and diagnose the root cause of these cable failures in the field was therefore undertaken. To determine which cable diagnostic technique to apply to a particular cable system, the type of cable insulation is an important criterion. Cables are classified into two main cable insulation groups:
Extruded/solid dielectric cable: These cables are whose insulation is extruded on the conductor and include cables, such as PE, XLPE, and EPR cables.
Laminated cable: These are cables whose insulation is made up of lami- nated layers, such as PILC cable.
The research investigating the premature failure of extruded dielectric insulated cables pointed to water tress and partial discharges (PDs) in the void cavity of the insulation as the main cause of these cable failures. Water trees are tree-like structures which, through a process of electrophoresis, grow and mature in extruded cables. Water trees do not occur in laminated insulated cables because these laminated cables do not have cavity voids as the extruded insulated cables.
The extruded solid dielectric cables are susceptible to voids during manu- facturing of these cables. After these cables are installed in the ground (i.e., in duct banks or direct buried), the voids over time will fill-up with water or water vapor. Therefore, water filled voids in the extruded insulation are referred to as water trees because these voids when examined under a microscope resemble like a tree, i.e., each void has a trunk and branches. Research has shown that water treeing is the most important form of degra- dation that may afflict older XLPE and high-molecular weight PE-extruded cables. As a result, the phenomenon of water treeing has been studied exten- sively, including means by which the degree of water tree-induced degrada- tion can be assessed. Water treeing can be described as a self-propagating dendritic pattern of electrooxidation, which reduces the AC and impulse breakdown strengths of extruded insulation and is the primary mechanism of degradation of extruded medium-voltage power cables. Although studied extensively, the initiation and growth mechanisms of water treeing are not clearly understood; they are not a single mechanism but complex interactions
of chemical, electrical, and mechanical phenomena that depend on the mate- rial and applied stresses. The visible manifestation of water treeing is strings of water-filled microcavities. The water-filled microcavities are connected by electrooxidized tracks, which are usually less than 0.1 μm in diameter, which is too small to see.
Water trees do not generate partial discharges (PD) by themselves. However water trees can lead to electrical trees as a result of a lightning impulse, or applied AC voltage, or during fault locating activities, or during DC high- voltage (HV) testing. The likelihood of causing a preexisting water tree to lead to an electrical tree may increase during a cable testing with high test voltages and the test duration. In general, electrical trees are more difficult to initiate than to grow, so that an electrical tree, once initiated, tends to grow to failure by PDs. Thus one can conclude that growing water trees do not gener- ate PD signals, unless they give rise to an electrical tree. Any PDs at a water tree imply the existence of one or more electrical trees at that water tree. In order for water trees to grow in extruded insulated cables, four factors need to be present in extruded cable insulation. These four factors are electrical field, time, water in void cavity, and entry point into the cable. Water trees slowly migrate across the insulation, ultimately bridging adjacent voids across the insulation of the cable. Literally thousands of these trees grow to form electrooxidized channels which are extremely small in diameter. Intuitively, as these water tree channels start to bridge the insulation, the losses dissipated through the insulation increases and thus lead to cable fail- ure over time.
This loss can be determined by measuring the dissipation factor (DF). Although other methods are available to determine the degree of water tree- ing in cables, the most widely used method is the measurement of DF (or PF) of the cable insulation. A perfect cable can be electrically modeled by a single capacitor. The longer the cable, the larger the capacitance of this capacitor. As water trees start to bridge the once-perfect cable insulation, this capacitor now starts to have some resistive (water tree) paths in parallel with it. The result is that the resistive loss component (in-phase component) of the total current loss increases which is measurable by measuring the DF or the PF of the cable. The DF readings (measurements) can be compared with previous test mea- surements and trended to assess the cable health. The reader should refer to Chapter 3 for further reading on DF and PF testing of an insulation system.
In performing a DF test, the applied voltage is usually stepped up from below operating voltage to slightly above operating voltage. Cables with poor insulation have higher DF (tan d) values than normal, and will exhibit higher changes in the tangent delta values with changes in applied voltage levels. Good cables have low individual tangent delta values and low changes in tangent delta values with higher applied voltages levels. In practice a very low-frequency (VLF) HV test is often used as the voltage excitation source to perform the tangent delta tests. VLF as an energizer source is preferred for two reasons: (1) the increased load capability in field applications in which 60 Hz is too bulky and expensive, and (2) the increased sensitivity and
effectiveness of measuring DF in the low frequency range for extruded cable. Tangent delta testing is also independent of the length of the cable, as it is a ratio of resistive losses to capacitive losses (the cable itself). Since XLPE and some EPR cables have very low tangent delta values when in good condition, the tangent delta resolution of the measurement equipment should be at least 1 × 10−4 to get accurate, meaningful results. In addition, a guard circuit to drain off surface leakage currents at the terminations should be used to give true tangent delta results during a measurement. This normally requires a VLF test instrument with a virtual ground return, instead of a solidly grounded VLF generator. Refer to Section 220.127.116.11 for detail discussion on VLF and tan d tests.
PD is defined as a localized electrical discharge that only partially bridges the insulation between two electrodes/conductors. It is important to note that this is a partial breakdown in the insulation of a cable and, therefore, would not be detectable using conventional fault location equipment. PD can occur from a number of locations within a cable system, such as within gas voids, in an electrical tree channel, along an interface (e.g., in a splice), between the concentric neutral to outer semiconducting layer, etc. When these PDs occur within the insulation section of XLPE cables, complete cable failure is imminent. During off-line field testing of cables with PD equip- ment, it is possible to elevate the applied voltage to detect one or multiple PD sites that may only discharge above certain voltage levels. The voltage at which a site starts to partially discharge is called the PD inception voltage (PDIV). If the PDIV values approach close to system-operating voltage levels, the cable will probably fail in service.
The erosion of the insulation by PD activity leads to what is referred to as an electrical tree. The PD and subsequent electrical trees rapidly lead to com- plete cable failure within XLPE cables. It should be noted, however, that some materials are more resistant to PD than others. Joints and terminations, for example, have a great ability, at least for a while, to fend off PDs in their insu- lation. Therefore, the location of the PD site is an important criterion to deter- mine whether that site will lead to imminent failure or not.
PD measurements on cables traditionally were performed by cable manu- facturers as a final quality control test. The PD tests are usually conducted in a shielded PD free test room. It is only within the last 5–10 years that advances in technology have allowed this diagnostic method to be used in the very noisy field environments. The ability to detect and locate sites of PDs down to 10 pC in cables in the field is now available. It should be noted that there are no PDs associated with water trees by themselves unless the water trees become electric trees. Therefore, unless water tree in the cable becomes an electrical tree (due to excessive electric stresses being present on the tree structure), PD testing is not able to detect it. Electrical trees and water trees have completely different properties, and the diagnostic methods used to detect them are also completely different. In many cases, cables with very poor DF test results, indicating the presence of severe water treeing, show no PD activity. PD is useful in finding installation defects in the cable system
and, in particular, in the accessories, however, PDs must be present in order to detect any PD. A wet splice may, for example, have a high leakage current but may not exhibit any PD. So, which method should be used to determine the health of the cable system? The diagnostic method applied will depend on a number of factors, including the age of the cable, type of insulation, mainte- nance strategy, etc. In order to diagnose the condition of a new installation, a PD test is very effective in locating installation defects that may have occurred. A poorly installed splice or an outer shield compromised during the installa- tion of the cable will lend itself more to a PD test than a tangent delta test, since no insulation aging (such as water trees) would be present in the new cable. For maintenance testing of older installations, a tangent delta test would be of most beneficial to determine the degree of insulation aging in the cable. If the cable is very critical in nature and even a single cable fault is to be avoided, then a combination of a PD and a tangent delta test is the best option.
Most utilities/cable owners are concerned about spending large amounts of unnecessary resources repairing cables that have a succession of repetitive failures. This is particularly true if the cable is globally deteriorated. The util- ities/cable owners would rather replace such a cable at the outset. In such a case, a tangent delta test will be most beneficial. Although it may not detect a singular defect in an otherwise good cable, it will detect a globally aged cable that could be the source of many future failures. As in most effective maintenance strategies, a combination of more than one diagnostic test is often the best way of establishing the condition of a cable system. Cable diag- nostic systems that include a combination of both tangent delta and PD diagnostic measurements in one integrated test instrument are now available to fulfill all these requirements.
Safety Practices and Grounding
When testing cables, personnel safety is of utmost importance. All cable and equipment tests shall be performed on isolated and de-energized systems, except where otherwise specifically required and authorized. Some switches may be connected to a cable end and serve to isolate the cable from the rest of the system. The ability of the switch to sustain the VLF test voltage while the other end is under normal operating voltage shall be checked with the manufacturer. The safety practices shall include, but not be limited to, the following requirements:
1. Applicable user safety operating procedures
2. IEEE std 510-1983 (reaffirmed in 1992)
3. NFPA 70E Standard for Electrical safety requirements for employee workplaces
4. Applicable state and local safety operating procedures
5. Protection of utility and customer property while testing, one or more cable ends will be remote from the testing site, therefore, before testing is begun, the following precautions shall be taken:
a. Cable ends under test must be cleared and guarded
b. Cables must be de-energized and grounded
c. At the conclusion of HV testing, attention should be given to dis- charge cables and cable systems including test equipment
d. Grounding requirements for cables and test equipment to elimi- nate the aftereffects of recharging the cables due to dielectric absorption and capacitance characteristics
Cable systems can be considered de-energized and grounded when a con- ductor and metallic shield are connected to system ground at the test site and, if possible, at the far end of the cable.
When testing, a single system ground at the test site is recommended. The shield or concentric conductor of the cable to be tested is connected to a system ground. If this connection is missing, deteriorated, or has been removed, it must be replaced at this time. A safety ground cable must connect the instrument case with the system ground. If the test instrument is a HV device, the safety ground cable should be at least a braided or stranded #2 AWG (34 mm2) copper cable capable of carrying available fault current. Only after the safety ground cable is in place, should the test cable be connected to the conductor and metallic shield; the conductor-to-ground connection shall now be removed. Should a local ground be advisable or required for the test equip- ment, the case ground must remain connected to the system ground in order to maintain an acceptable single-ground potential. Care should be taken to ensure that all ground connections could not be disconnected accidentally.
Cable Testing Methods
After a new cable has been installed and before it is energized, acceptance proof testing (HV tests) should be performed. In general, acceptance proof test are conducted at 80% of final factory test voltage. Also, routine mainte- nance HV tests may be conducted in the field on installed cables as maintenance tests. The maintenance HV tests are conducted at 60% of final factory test voltage. The following tests may be performed in the field for acceptance and maintenance of cables.
Insulation Resistance and DC Hi-Pot Testing
In the past, insulation resistance and DC HV (hi-pot) tests have been used for acceptance (proof) and maintenance testing of cables. When testing cables with DC voltage, it should be understood that DC voltage creates within the cable insulation system an electrical field determined by the conductance and the geometry of the cable insulation system. However, the normal service voltage applied to cable is AC 60 Hz voltage, thus during normal service conditions the AC voltage creates an electrical field that is determined by the dielectric constant (capacitance) of the insulation system. Therefore the electric stress distribution with DC voltage will be different than with AC voltage. Further, conductivity is influenced by temperature to a greater extent rather than the dielectric constant, therefore comparative electric stress distribution under DC and AC voltages will be affected differently by changes in temperature in the insulation. The DC voltage tests are effective in detecting failures that are triggered by thermal mechanism. The value of the DC voltage diagnostic tests for extruded-type insulation are somewhat limited because failures under service AC voltage con- ditions are most likely to be caused by PDs in the voids of extruded insulation rather than by thermal mechanism. On the other hand, the DC voltage diagnos- tic tests are very meaningful for laminated-type insulation system where the failure is most likely to be triggered by thermal mechanism. The current trend is to minimize the use the DC hi-pot tests on extruded insulation for the reasons discussed above and because of potential adverse charging effects of DC hi-pot tests on extruded insulation. The reader should refer to Section 2.5 for perform- ing insulation resistance and the DC hi-pot tests on cables.
AC Hi-Pot Testing
Cables and accessories may also be field tested with 60 Hz AC voltage, although this is normally not done because of the requirement for heavy, bulky, and expensive test equipment that may not be readily available or transportable to a field site. The most common field tests performed on cables are DC hi-pot or VLF tests, such as one-tenth of hertz frequency tests in lieu of AC hi-pot tests. However, if AC hi-pot acceptance and mainte- nance tests are to be conducted on cables, then it should be borne in mind that this test is not very practical in the field. Further, the AC hi-pot test can only be conducted as go-no-go test, and therefore it may cause extensive damage should the cable under test fails, i.e., a disruptive discharge through the insulation takes place during the test. On the other hand, AC hi-pot test has a distinct advantage over other test methods of stressing the insulation comparably to normal operating voltage. Further, this test replicates the fac- tory test performed on the new cable. When performing the AC 60 Hz hi-pot test consideration should be given to the adequacy of the test equipment for successfully charging the test specimen. The AC test equipment should have adequate volt-ampere (VA) capacity to supply the required cable charg- ing current requirements of the cable under test. The VA capacity of the AC hi-pot test equipment may be determined by the following formula.
The test voltage values recommended for acceptance and maintenance tests are 80% and 60%, respectively, of the final factory test voltage. The test connections are similar to the connections indicated in Section 2.5, for DC testing of cables.
PF and DF Testing
PF and DF may be performed on shielded cable systems to determine insulation degradation to reduce in-service cable failures. The PF tests for shielded or sheathed cables and accessories are discussed in Section 3.6.8. These tests are described as diagnostic testing techniques for field testing of service- aged cable systems. For lossless insulation, the cable capacitance (C) per unit length can be defined by the following equation:
When a voltage V is applied to the loss-free insulation system (dielectric), the total current IT drawn by the dielectric is the sum of the capacitive charging current IC and loss current (resistive) IR. As was discussed in Section 3.2.2, Figure 3.1, the angle formed by the current IT and IC is d, and the angle formed by the IT and voltage E is q where cos q is the PF of the dielectric. The DF (tan d) test allows an evaluation of an insulation system at operating voltage level and frequency. The tan d test can also be performed at frequency other
than 60 Hz, such as at VLF of 0.1 Hz during proof test conducted at such frequency. According to IEEE std 400-2001, tests conducted on 1500 miles of XLPE insulated cable have established a figure of merit for XLPE at tan d =
2.2 × 10−3. If the measured tan d is greater than 2.2 × 10−3, then the cable insula- tion is degraded by moisture in the form of water trees (voids in the insula- tion filled with water), and it is recommended that additional hi-pot tests, such as VLF test be conducted to identify the defects in the cable insulation. The tan d test for each conductor with respect to ground should be made. The evaluation should be based upon comparative analysis with previous test results or correlated with test results of similar types of cables.
Very low frequency (VLF) test is conducted with an AC voltage at frequency ranging from 0.01 to 1 Hz. VLF test can be classified as withstand or diagnostic test, i.e., it may be performed as a proof test for acceptance or as a maintenance test for assessing the condition of the cable condition.
For the withstand test, the insulation under test must withstand a speci- fied applied voltage that is higher than the service voltage across the insulation for a specified period of time without breakdown of the insula- tion. The magnitude of the withstand voltage is usually greater than that of the operating voltage. If the VLF test is performed as a diagnostic test, it is performed at lower voltages than withstand tests, and therefore may be considered as nondestructive test. Diagnostic testing allows the determi- nation of the relative amount of degradation of a cable system, and by com- parison with previous test data, whether a cable system is likely to continue to perform correctly in service. It should be noted that values of the diagnostic quantity measurements obtained during VLF tests may not cor- relate with those obtained during power frequency tests. For example, the PF and DF tests are conducted at power frequency (60 Hz) which is much higher than at 0.1 Hz, and PD may differ in terms of magnitude and incep- tion voltage. When a cable system insulation is in an advanced condition of degradation, the VLF diagnostic tests can cause breakdown of the cable before the test can be terminated. The VLF withstand test for cable systems may be conducted with the following waveforms:
1. With cosine-rectangular waveform
2. With sinusoidal waveform
3. With bipolar rectangular waveform
4. With alternating regulated positive and negative DC step voltages The diagnostic test using VLF methods for cable systems are
• VLF dissipation factor (tan d) measurement (VLF-DF)
• VLF differential dissipation factor measurement (VLF-DTD)
• VLF dielectric spectroscopy (VLF-DS)
• VLF loss current harmonics (VLF-LCH)
• VLF leakage current (VLF-LC)
• VLF partial discharge measurement (VLF-PD)
The most commonly used, commercially available VLF test frequency is
Hz. Other commercially available frequencies are in the range of 0.0001–
1 Hz. These frequencies may be useful for diagnosing cable systems where the length of the cable system exceeds the limitations of the test system at 0.1 Hz, although there is evidence that testing below 0.1 Hz may increase the risk of failure in service following the test. The internal impedance of the test set can limit the available charging current, preventing the cable under test from reaching the required test voltage. Cable manufacturer may be consulted when selecting an initial test voltage level and testing time duration for a particular cable length. VLF test voltages with cosine-rectangular and the sinusoidal wave shapes are most commonly used. While other VLF wave shapes are available for testing of cable systems, recommended test voltage levels have not been established.
During a VLF test an electrical tree at the site of an insulation defect is forced to penetrate the insulation. Inception of an electrical tree and channel growth time are functions of test signal frequency and amplitude. For an electrical tree to completely penetrate the insulation during the test duration, VLF test voltage levels and testing time durations have been established for the two most commonly used test signals, the cosine-rectangular and the sinusoidal wave shapes.
The voltage levels (installation and acceptance) are based on most-used practices worldwide of between two times rated voltage and three times rated voltage for cables rated between 5 and 35 kV. The maintenance test level is about 80% of the acceptance test level. One can reduce the test voltage another 20% if more test cycles are applied. Tables 6.4 and 6.5 list voltage levels for VLF withstand testing of shielded power cable systems using cosine-rectangular and sinusoidal waveforms. For a sinusoidal waveform, the rms is 0.707 of the peak value if the distortion is less than 5%.
The recommended testing time varies between 15 and 60 min, although the average testing time of 30 min is usually used. The actual testing time and voltage may be defined by the supplier and user and depend on the testing philosophy, cable system, insulation condition, how frequently the test is conducted, and the selected test method. When a VLF test is interrupted, it is recommended that the testing timer be reset to the original time when the VLF test is restarted.
The tan d may be performed with VLF equipment at 0.1 Hz sinusoidal to monitor the aging and degradation of extruded insulated cables. The tan d test provides an assessment of the water tree damage in the cable insulation.
The tan d measurement with 0.1 Hz sinusoidal waveform provides comparative assessment of the aging condition of PE-, XLPE-, and EPR-type insulation
• The 0.1 Hz cosine-rectangular waveform has polarity changes similar to those at power frequency. Because of the sinusoidal transitions between the positive and negative polarities, traveling waves are not generated, and because of continuous polarity changes, dangerous space charges are less likely to be developed in the insulation.
• Leakage current can be measured.
• Cables may be tested with an AC voltage approximately three times the rated conductor-to-ground voltage with a device comparable in size, weight, and power requirements to a DC test set.
• The VLF test can be used to test cable systems with extruded and laminated dielectric insulation.
• The VLF test with cosine-rectangular/bipolar pulse and sinusoidal waveform works best when trying to locate a few defects from oth- erwise good cable insulation.
• When testing cables with extensive water tree degradation or PDs in the insulation, low frequency withstand testing alone may not be conclusive. Additional diagnostic tests that measure the extent of insulation losses will be necessary.
• Cables must be taken out of service for testing.
18.104.22.168 PD Test
A PD is an electrical discharge that occurs in a void within the extruded cable insulation, at interfaces or surfaces such as shield protrusion and the insulation, or in a water tree within cable insulation when subjected to moderately HV. PD occurs as a series of PD pulses during each half cycle of an AC waveform. The rise time of the PD pulses is in the order of nanoseconds to tens of nano- seconds. The PD pulses tend to set an electromagnetic field which propagates in both directions along the cable with a velocity of propagation based on the dielectric constant of cable insulation. PD characteristics depend on the type, size and location of the defects, insulation type, voltage, and cable tempera- ture. The insulation of full reels of extruded cables is tested for PDs at the factory at power frequency. This test is known to detect small imperfections in the insulation such as voids or skips in the insulation shield layer. It seems logical to perform PD measurements on newly installed and service-aged cables to detect any damage done during the installation of new cable or in-service degradation of cable insulation due to PDs.
There are two approaches that can be used for detecting PDs from installed cables in the field. They are on- and off-line detection system. There are sev- eral commercial off-line systems available for measuring PD in medium- voltage systems (up to 35 kV). The online measuring system is based on measuring PDs at the cable-operating voltage. On the other hand, in the off- line system the PD measurements are done at a higher voltage than cable- operating voltage. This is due to the fact that the off-line testing requires the cable to be de-energized which results in cessation of any active PD activity. In order to activate the PD activity again in the de-energized cable during off-line testing, a higher voltage is needed to reinitiate the PD activity. The test instruments for PD testing for online or off-line comprise of the power supply, sensors and noise reduction, signal detection, and signal processing equipment. The power supply can be 60 Hz voltage, oscillating voltage, or VLF (0.1 Hz) voltage source. The sensors can be inductive couplers, capacitive couplers, or an antenna along with noise treatment and amplification equip- ment. The signal detection and processing equipment includes digital oscil- loscope, spectrum analyzer, wave form digitizer, time-domain reflectometer (TDR) (time resolved) and/or frequency resolved.
Although it is difficult to conduct a PD measurement in the field because of external noise, this test can be performed in the field where conditions warrant it is worth the time and expense to do so. The PD test gives the most convincing evidence whether the cable is in good condition and suit- able for service or needs to be repaired or replaced. The PD test is useful for both the laminated and extruded cable insulation systems. This test can be performed at power frequency or at any other frequency, such as Hz (VLF).
To perform an off-line PD test the cable is disconnected from the network at both ends and correctly isolated. A voltage source and a coupling device, or sensor, are connected at one of the ends (near end), whereas the remote end is left open. The coupling device could be capacitive or inductive. The coupling device is connected to the PD detecting and processing systems.
Variations of this setup include a measuring system with sensors at both ends and means to communicate the far end data to the near end processing devices or, in the case of a branched system, sensors placed at the end of each branch. Multiterminal testing also has the advantage of greater sensitivity in the PD testing of very long cable lengths as the pulse travel distances are considerably shorter and consequently the related attenuation of pulse amplitude will be less. The following steps are implemented:
1. Low-voltage TDR is used to locate cable joints (splices) and other irregularities
2. Sensitivity assessment
3. PD magnitude calibration
4. PD testing under voltage stress
5. Test data analysis and documentation
The purpose of this step is to determine the value in picoCoulomb (pC) of the smallest PD signal detectable under the test conditions. In extruded dielectric cables PD activity in the range of several pCs is required, otherwise inadequate detection sensitivity may mask the existence of serious defects with low PD magnitudes. Inability to detect low levels of PD may result in false-negative situations that are expected to lead to unexpected post-testing service failures. In addition incorrectly identified PD may lead to false-positive situations leading to unnecessary cable replacement.
Therefore, a calibrated pulse, such as 5 pC, is injected at the near end. The PD estimator detects and records the response. If the reflected signal cannot be seen above the filtered noise level, a larger signal, such as 10 pC, is injected.
This process is repeated until the reflected signal is observable. This deter-mines the smallest PD signal that can be resolved under the test conditions.
PD magnitude calibration
The calibrated pulse generator is connected to the cable remote end. A large signal, such as 50 or 100 pC, is injected. The corresponding signal recorded at the near end is evaluated by integrating it with respect to time. The constant k is adjusted until the PD magnitude read is 50 or 100 pC. The instrument is now calibrated for measuring the apparent charge, q, of the PD.
PD testing under voltage stress
Off-line tests can be carried out using different voltage sources. There is a good technical basis for testing up to 1.5 to 2.5 times of rated voltage to ensure that the PDIV of the cable is sufficiently high enough to activate the PD activity. There is an increased risk of initiating damage at defects in aged cable systems that are innocuous at operating voltage if testing is carried out at voltages greater than 2.5 times of rated voltage. There is also an increased risk of failures during the PD testing. However, some utilities will request testing up to a maximum of three times of rated voltage on new cables, either on the reel or newly installed, to ensure that there was no damage during transportation or installation. In addition, some utilities will test up to three times of rated voltage, even though there is a significantly higher probability of failure during the testing, of the following cable systems:
• Cable circuits with generic defects that may cause high failure rates, e.g., some silane-cured cables can cause severe corrosion of alumi- num conductors.
• Cable circuits that are being considered for silicone injection, the rationale being that all cables with electrical trees will fail at higher test voltages. The higher test voltages could also initiate new electrical trees.
• Cable circuits that may have suspect accessories and/or cables to ensure operation during high load periods, e.g., during the summer months in some urban areas. The voltage in power frequency tests
may be applied for up to a maximum of 15 min to ensure that elec- trons are available in cavities to initiate PD. However, once PDs are detected, the voltage should be applied long enough to collect suffi- cient data up to a maximum of 15 s. Some PD testing organizations will decrease the voltage very soon after the onset of steady PD when testing extruded dielectric cable circuits.
As an example, the following steps are conducted for testing for voltage stress. The voltage is rapidly raised to the cable operating level (1.0 p.u.) at which it is maintained for several minutes as a conditioning step. The voltage is ramped to its maximum value (such as 2.0 or 2.5 p.u.). It then is returned to zero as quickly as possible. During this stress cycle, several sets of data are captured, each set encompasses an entire 60 Hz period. The rising and falling parts of the voltage help determine the PDIV and PD extinction voltage (PDEV), respectively. It should be apparent that off-line testing using higher voltages (elevated stress) than cable operating voltage may be a destructive test.
In summary, it is not possible to standardize a specific test protocol at the current time for either online or off-line tests. This may become possible as more data are collected. For off-line tests, the amplitude of the test voltage can be varied. For heavily aged systems, a maximum test voltage of 2 p.u. is suggested. As the anticipated condition of the cable improves, the test volt- age may be increased to as much as 2.5 p.u. New cables, either on the reel or newly installed, may be tested to a maximum of 3 p.u. at the concurrence of the cable owner and cable manufacturer. The test duration should be long enough to allow the availability of electrons to initiate PDs, but once PDs are detected, the voltage should be applied long enough to collect sufficient PD data.
PD test data analysis and documentation
The PD test provider should provide cable users with a report of the cables tested and the PD test results. The PD test provider should give the cable user recommendations on possible corrective action to be taken. The report of the test results should include the value of PD detection sensitivity and a reference to the method used in obtaining this value. The PD site location results must also be provided with an assessment of the accuracy limits within which these results can be interpreted under the conditions of the specific test. This becomes critical where the location is at or near a splice. Details to be included in the report are as follows:
Cable system identification
1. Name of cable manufacturer
2. Cable section identification (i.e., substation name, from switch number to switch number)
3. Cable voltage class
4. Cable insulation (if mixed, specify)
5. Operating voltage (phase-to-neutral)
6. Conductor type and size (if mixed, specify)
7. Cable length
8. Location of splices
9. Cable vintage or year placed in service
10. Neutral type, for example, concentric wires, metal tapes, or flat strap, and size
11. Type of construction, i.e., direct buried, duct, aerial, jacketed, unjack- eted, and so on
12. Splice type, if available
13. Termination type, i.e., pole-top, switching cabinet live-front/dead front, premolded, heat-shrink, and so on
PD test results
1. Test date.
2. Date of the most recent previous test.
3. Estimated cable length.
4. Splice location.
5. Background noise level.
6. Minimum resolvable PD signal pC magnitude (sensitivity) and how it was determined. If the sensitivity is lower than expected, provide the reason(s).
7. Test voltage levels.
8. At each test voltage level, the location of each PD site, along with the limits of accuracy.
9. At each voltage and site location, the number of PD events per second or per cycle of a sinusoidal excitation voltage.
10. At each voltage and site location, a phase-resolved PD representation (pC vs. phase angle for each PD event recorded), provided the excita- tion voltage is sinusoidal. Specify the number of cycles included in the phase-resolved diagram.
11. For a frequency-domain measurement, describe the spectral characteristics and the estimated location for each PD site. Specify the limits of accuracy.
12. Any other diagnostic results pertinent to the test method used.
13. An indication of the severity of the PD behavior, if PDs are detected, and recommendations on possible corrective action to be taken.
14. The format of data reporting may vary. For instance, some prefer reporting individual PD events in a three-dimensional (3D) form
with location, pC level, and phase angle at which each PD is initiated.
15. Variations of this 3D representation are also possible. Others prefer a set of two-dimensional representations, showing PD location with PDIV, and apparent charge (pC) versus phase angle for each PD site, at each voltage level, and PD repetition rate for each PD site at each voltage level.
AC Resonance Test
The resonant test systems are used to test cable and other electrical apparatus with AC voltage at power frequencies (50 or 60 Hz). This method has the advantage over other test methods, of stressing the insulation similar to normal operating conditions. In the past to test electrical apparatus at power frequency required bulky and expensive test equipment that was not porta- ble for on-site field testing applications. The resonant test systems were developed that can be handled easily on-site for testing. Since the mid-1990s, the resonant test systems have been used for testing medium and HV cables in Europe and United States. This method can be used to test cable consisting of either XLPE, oil-impregnated paper and EPR, or a combination of these insulating materials. As the name implies, this test method is based on using AC at the operating frequency (50 or 60 Hz) as a test source using the princi- ple of resonance. Resonance is defined as the condition at which the net inductive reactance cancels the net capacitive reactance at operating fre- quency. The resonant circuit must have both capacitance provided by the cable under test and inductance provided by the reactor of the test set. When resonance occurs, the energy absorbed at any instant by one reactive element is exactly equal to that released by another reactive element within the system. In other words, energy pulsates from one reactive element to the other. Therefore once the system has reached a state of resonance, it requires no further reactive power since it is self-sustaining. The total apparent power is then simply equal to the average power dissipated by the resistive ele- ments in the inductor and cable system. Either parallel or series resonant circuits are normally used for conducting this test.
The series resonant test consists of a voltage regulator of an autotrans- former type (Toroidal, CTVT or Thoma) is connected to the supply volt- age. The regulator provides a variable voltage to the exciter transformer. The exciter transformer is fed by the output of the voltage regulator. This transformer steps the voltage up to a usable value by the HV portion of the circuit. The HV reactor L and the load capacitance C represent the HV por- tion of the circuit. The inductance of the HV reactor can be varied by changing the air gap of the iron core. The load capacitance C consists of the capacitance of the load. The coupling capacitance for PD measurement, stray capacitance and, in the case of tank-type (T) sets, the HV bushing. When testing, the HV reactor is adjusted so that the impedance of L corresponds to the impedance of C at the frequency of the supply voltage. Thus the circuit is tuned to series resonance at 50 or 60 Hz. The Q of the basic resonant circuit or with a low loss test specimen (e.g., XLPE cable, sulfur hexaflouride switchgear, bushing, etc.) is typically 50 to 80. The HV reactor is designed for a minimum Q of 40. The system Q is designed around the anticipated load. In case of a flashover during testing on the HV side, the resonant circuit is detuned and the test voltage collapses immediately. The short-circuit current is limited by the impedance of the HV reactor. This means that the short-circuit current of a series resonant system with a Q of 40 is 2.5% of the load current to which it is tuned. The series resonant mode is well suited for sensitive PD measurements. Harmonics from the supply are better suppressed than in parallel mode.
The parallel resonant mode provides a more stable output voltage with test specimens, such as large generator windings, or other specimens with corona losses. The test voltage rate of rise is stable in parallel mode, independent of the degree of tuning and the Q of the circuit. Furthermore, parallel mode allows the test set to be energized to full voltage without a load. This is useful for calibrating the instrumentation and checking for the PD level of the test equipment. The test voltage rate of rise is stable in parallel mode, independent of the degree of tuning and the Q of the cir- cuit. Furthermore, parallel mode allows the test set to be energized to full voltage without a load. This is useful for calibrating the instrumentation and checking for the PD level of the test equipment. The average power absorbed by the system will also be at a maximum at resonance. The com- monly used measure of the quality in a resonant circuit is the quality factor, or Q. The power source of resonant circuits operating in the reso- nant mode (exciter and regulator) is used to supply the dissipated energy. Q is approximately equivalent to the ratio of the output kVA to the input kVA. Given kVA requirements of the load and the Q of the test system, the input power can be obtained by dividing the kVA by the Q. The correct mode of operation must be chosen according to the test objects and the measurements to be carried out. The parallel resonant mode provides a more stable output voltage with test specimens, such as large generator windings, or other specimens with corona losses.
Resonant test system are available that use variable inductance and vari- able frequency resonant and pulsed resonant test sources. A brief descrip- tion of the variable frequency resonant test system is as follows. The resonant test system with variable frequency mainly consists of the fre- quency converter, the exciting transformer, the coupling capacitors, and HV reactors with fixed inductance. The frequency converter generates a variable voltage and frequency output which is applied to the exciter transformer. The exciter transformer excites the series resonant circuit consisting of the reactor’s inductance L and the cable capacitance C. The resonance is adjusted by tuning the frequency of the frequency converter according the formula:
The average power absorbed by the system will also be at a maximum at resonance. The commonly used measure of the quality in a resonant circuit is the quality factor, or Q. The power source of resonant circuits operating in the reso- nant mode (exciter and regulator) is used to supply the dissipated energy. Q is approximately equivalent to the ratio of the output kVA to the input kVA. Given kVA requirements of the load and the Q of the test system, the input power can be obtained by dividing the kVA by the Q. The correct mode of operation must be chosen according to the test objects and the measurements to be carried out. The series resonant mode is well suited for sensitive PD measurements as well.
Summary of Testing Methods
The purpose of summarizing cable testing methods is to cite the advantages and disadvantages of these tests so that the reader can quickly determine the test method best suited for his application. The cable tests can be categorized into three categories: (1) hi-pot withstand tests, (2) general condition assessment (GCA) tests, and (3) PD tests. These tests can further be viewed from the perspec- tive of being destructive or nondestructive tests. Any test that uses the test source voltage to be higher than the in-service operating voltage could be classified as destructive test because during testing the cable insulation will be subjected to a higher voltage than what it will see in service. Therefore, all hi-pot withstand tests would fall into this category. However, during a hi-pot test, if the voltage is applied in a steps and the leakage current is monitored, then the test may be clas- sified as being nondestructive. The reasoning for this is that the test can be aborted before the insulation gets to a failure point since at every step of voltage application the leakage current is being monitored and evaluated before pro- ceeding to the next step. An application of this test procedure is the step-voltage DC hi-pot withstand test. The same cannot be said of AC hi-pot withstand test since there is no way to evaluate the leakage current, therefore this test would be considered as go-no-go test and considered to be destructive. The GCA tests and PD tests are classified as nondestructive since the voltage applied during these tests is either the same, or lower than, or slightly above the in-service operating voltage. The advantages and disadvantages of the tests are as follow:
Hi-pot withstand tests
Under this category, cable tests that use HV source are listed. These tests are:
DC hi-pot tests, AC hi-pot, AC resonant test, and VLF test. The advantages and disadvantages of the test that use hi-pot voltage source are DC hi-pot test:
1. Has a long history of use
2. Very portable and convenient for field test
3. Low power requirements
4. Is a good for conductive type defects (water in laminar cables)
• Demonstrated to induce space charge which aggravates defects in aged extruded cable long after the test’s conclusion
• Is blind to high impedance defects such as voids and cuts
• Stress distribution is not the same as in-service conditions
• Cannot be compared to factory tests 60 Hz hi-pot test and AC resonant test
• Is good for conductive and high impedance defects
• Does not induce space charge, thereby minimizes the propagation of defects in extruded cable
• Replicates steady-state in-service conditions
• Can be compared to factory tests
• Most expensive and not practical for field tests
• Highest power requirement except for AC resonant test
• Grows certain type of defects VLF hi-pot tests
• Portable for field testing
• Relatively low power requirements
• Is a good for conductive-type defect and high-impedance defects
• Does not induce as much space charge as DC hi-pot in aged extruded cable
• Causes some defects to grow rapidly resulting in shorter test time
• Aggravate defects in aged cable without failing them
• Does not replicate service conditions
• Cannot be directly compared to factory tests
• Not recommended for aged cable with multiple defects
• Stress distribution is not the same as in-service conditions
• Does not replicate normal stress distribution conditions with wet regions
GCA tests are those which determine the overall health of the cable insulation. These tests include: DF/tan d/PF, PD tests, dielectric spectroscopy, depolarization-return voltage, and depolarization-relaxation current. Each of these tests has their own advantages and disadvantages. Some of these tests are not covered in this book since they are beyond the scope. In general, the following can be stated for PF/DF and PD tests.
PF/DF (tan d)
• Considered to be nondestructive to cable insulation
• Tests are conducted at in-service voltage levels
• Monitor the overall condition of the cable insulation
• Effective in detecting and assessing conduction-type defects
• Can be compared to factory tests
• Portable for field testing
• Require prior cable types and data for comparison
• Temperature dependant in extruded cables
• Blind to high-impedance defects such as cuts, voids, and PD
• Cannot detect singular defects in extruded insulation, such as water tree
• Not an effective test for mixed dielectric or newly installed cable
• Equipment is costly compared to hi-pot equipment
Two type of PD tests are considered, that is online PD testing and off-line PD testing. PD diagnostics tests are considered to be effective in locating defects in shielded power cables.
PD diagnostics tests
• Considered to be nondestructive
• Can detect and locate high-impedance defects such as void, cuts, electrical trees, and tracking
• Can be performed online in limited applications
• Effective at locating defects in mixed dielectric systems
• Limited to cables with a continuous neutral shield
• Requires a trained analyst to interpret measurements
• Cannot detect or locate conduction-type defects
• Not effective for branched network applications
Online PD diagnostics
• Performed while circuit is energized
• Detect and locate some accessory defects and some cable defects
• Does not require an external voltage source
• Detects only 3% or less of cable insulation defects in extruded cable
• Not a calibrated test, therefore the test results are not objective
• Cannot be compared to factory tests
• Not effective by statistically significant data correlating results to actual cable system defects or failures
• Requires access to the cable every few hundred feet depending on the cable construction
• Requires that manholes be pumped to access cable
• Cannot be applied to long directly buried cables
Off-line PD diagnostics
• Can be readily compared to factory baseline tests
• Replicates steady-state and transient operating conditions
• Can locate electrical trees with PDs
• Locates all defect sites in one test from one end of the cable
• Is effective with mixed dielectric cables
• Can test up to 1 to 3 miles of cable depending on the cable construction
• Provides onsite report of the test results
• Need circuit outage for test to be performed
• Equipment is expensive when compared to other tests