Power Cable Technical Analysis

Overview of design and construction

AC cables may be single-core (one current-carrying phase conductor) or multi-core.LV distribution cables are invariably multi-core, having 2–5 conductors including the neutral. MV cables are single-core or 3-core, many factors being likely to influence choice. Not least of these is conductor size (cross-sectional area). 3-core cables with conductors larger than 300 mm 2may be impracticable in view of their weight, inflexibility and difficulty in jointing and terminating. The growth in power demand implies a trend towards single-core cables for MV distribution, with conductors as large as 1 200 mm 2 not uncommon. The largest conductors are found in HV cables and the majority of modern HV cables are single-core, with conductor sizes typically up to 2500 mm 2.

The following figures illustrate the structure and principal components of some common cables, both legacy paper-insulated and modern extruded polymeric insulated constructions. Successive sections of this paper include details of component materials, their functions and a summary of manufacturing processes.

Fig 1 shows an LV 4-core armoured cable typical of use in industrial applications. Cables of this type meet BS 5467 [with polyvinyl chloride (PVC) oversheath] or BS 6724 (with oversheath having low emission of smoke and corrosive gas). Cables to these standards are also suitable for use at 3.3 kV ( U m = 3.6 kV).

Fig 1: Low voltage 4-core cable with sector-shaped stranded copper conductors, XLPE insulation, steel wire armour and extruded oversheath

Components of Fig 1 cable:

1. Stranded copper conductors (three-phase conductors + neutral).
2. Cross-linked polyethylene (XLPE) insulation.
3. Galvanised steel wire armour.
4. Oversheath of PVC or a polymeric material having low emission of smoke and corrosive gas.

Fig 2 shows an LV ( U m = 1.2 kV) 3-core CNE waveform cable typical of use by utilities. Cables of this type meet BS 7870-3.11 (with PVC oversheath) or BS 7870-3.12 (with oversheath having low emission of smoke and corrosive gas).

Fig 2: Low voltage waveform CNE cable with sector-shaped solid aluminium conductors, XLPE insulation, copper wire neutral/earth conductor and extruded oversheath

Components of Fig 2 cable:

1. Sector-shaped solid aluminium phase conductors.
2. XLPE insulation.
3. Bedding layer.
4. Concentric waveform copper wire neutral/earth conductor.
5. PVC or low-smoke and corrosive gas oversheath.

Fig 3 shows an 11 kV 3-core paper insulated belted cable with corrugated aluminium sheath and PVC oversheath, commonly installed by utilities before the adoption of extruded polymeric cables. This type of cable is known by the acronym PICAS (paper insulated corrugated aluminium sheathed). The term belted refers to a construction in which a conductive earth screen (component 5) is applied over the three laid up cores rather than around individual cores. Electrically the belted design is weaker than the alternative screened construction, and for this reason the more highly stressed 33 kV paper-insulated cables are all screened design.

Fig 3: 11 kV 3-core PICAS cable with sector-shaped stranded aluminium conductors, MIND belted paper insulation, corrugated aluminium sheath and PVC oversheath

Components of Fig 3 cable:

1. Sector-shaped stranded aluminium phase conductors.
2. Conductive paper screen.
3. Mass-impregnated non-draining (MIND) paper insulation.
4. Belt insulation papers.
5. Conductive paper insulation screen.
6. Corrugated aluminium sheath.
7. Bitumenised sealant layer.
8. PVC oversheath.

Fig 4 shows an 11 kV 3-core paper insulated screened cable with lead sheath, steel wire armour and hessian serving. These cables, and the alternative belted construction, were standard cables for industrial applications and were also installed by utilities. The descriptive acronym for this cable is PILC SWA (paper-insulated lead covered and steel wire armoured).

Fig 4: 11 kV 3-core PILC SWA cable with circular stranded copper conductors, MIND screened paper insulation, lead sheath, steel wire armour and hessian serving

Components of Fig 4 cable:

1. Circular stranded copper phase conductors.
2. Conductive paper screen.
3. MIND paper insulation.
4. Metallised paper insulation screen.
5. Circularising fillers.
6. Lead sheath.
7. Bitumenised hessian bedding.
8. Galvanised steel wire armour.
9. Bitumenised hessian serving with whitewash coating.

Fig 5 shows a 33 kV single-core cable with XLPE insulation, copper wire earth screen and medium density polyethylene (MDPE) oversheath. Cables to this design with rated voltages 11 and 33 kV meet BS 7870-4.10 and are typically installed by utilities. They are also available with a sheath having low emission of smoke and corrosive gas. A variant of this 33 kV cable has a lead sheath under the MDPE oversheath and is specified in BS 7870-4.11. Cables of the same basic design but with aluminium wire armour in place of the copper wire screen are commonly used in industrial applications.

Fig 5: 33 kV single-core cable with stranded copper conductor, XLPE insulation, copper wire screen and extruded oversheath

Components of Fig 5 cable:

1. Stranded copper conductor.
2. Extruded conductive screen.
3. XLPE insulation (thickness 8 mm for 33 kV).
4. Extruded conductive screen (semicon layer).
5. Copper wire earth screen.
6. MDPE or low-smoke and corrosive gas oversheath.

Fig 6 shows an 11 kV 3-core cable with XLPE insulation, copper tape earth screens, steel wire armour and MDPE oversheath. Armoured cables of this type with rated voltages 11 and 33 kV are specified in BS 6622 and are typically installed in industrial networks. Similar cables, to BS 7835, are available with sheaths having low emission of smoke and corrosive gas.

Fig 6: 11 kV 3-core cable with stranded copper conductors, XLPE insulation, copper tape screens, steel wire armour and extruded oversheath

Components of Fig 6 cable:

1. Stranded copper conductor.
2. Extruded conductive screen.
3. XLPE insulation (thickness 3.4 mm for 11 kV).
4. Extruded conductive screen (semicon layer).
5. Copper tape earth screen.
6. Circularising fillers.
7. Bedding sheath.
8. Galvanised steel wire armour.
9. PVC or MDPE oversheath.

Through the range of MV and HV rated voltages the major design features are the same, the main differences being in the choice of metallic screening and outer protection. Primary insulation thickness must, of course, must be appropriate to the rated voltage of the cable. Insulation thicknesses for MV cables are defined in standards and have remained unchanged for many years. The standard thickness for each U m is the same across the range of International Electrotechnical Commission (IEC), CENELEC and associated national standards.

The variables determining the electric stress E in the cable insulation for each rated voltage are the diameter over the conductor screen and the insulation thickness. Stress at a radial position r within the insulation is given by

E=V/r⋅ln[r2/r1]E=V/r⋅lnr2/r1

where, V is the phase-earth rated voltage U 0; r 1 is the outer radius of the conductor screen and r 2 is inner radius of the insulation screen.

Maximum stress is at the surface of the conductor screen. This is the design stress of the cable. For an 11 kV cable ( U 0 = 6.4 kV) with 185 mm 2 conductor, the conductor screen diameter will be about 20 mm. With standard insulation thickness 3.4 mm the stress at working voltage is 2.2 kV/mm. This is a very modest stress for a dielectric such as XLPE which has a test breakdown strength many times this value. A short length of 11 kV XLPE cable subjected to a short-time test to breakdown might be expected to withstand up to about 80 kV test voltage, equivalent to a maximum stress in the insulation of 28 kV/mm. What has to be taken into account, however, is the very large volume of insulation in a cable network, and the length of time (hopefully at least 40 years) during which the insulation will be under stress. Both these factors influence statistical breakdown probability, which users will require to be as low as possible and within an acceptable limit.

For cables of higher rated voltage, design stresses increase. For example, the stress in a 185 mm 2 33 kV cable with standard insulation thickness 8 mm is 3.2 kV/mm. For HV cables the insulation thickness is graduated according to conductor size, to take account of the fact that maximum stress is higher for smaller conductor diameters. One manufacturer's data sheet for 132 kV cables specifies insulation thickness 18 mm for a 500 mm 2 cable. In this case the maximum stress is 6.6 kV/mm. One practical reason for accepting higher stresses in higher voltage cables is the physical size of the cable if stresses as low as 2.2 kV were maintained through the voltage range. For example, a maximum stress of 2.2 kV in the 132 kV cable necessitates an insulation thickness of 150 mm, which is clearly a practical nonsense. From the statistical point of view, the total length of installed 132 kV cable is many times less than that of 11 kV, which to some extent mitigates the effect of increase in operating stresses at higher rated voltages.

Standards

International standards for cables and their accessories are developed and published by the IEC and CENELEC, the European Committee for Electrotechnical Standardisation. Work to develop CENELEC and IEC standards is done by groups of experts from member countries.

CENELEC standards are published as European Standards (prefix EN) or Harmonisation Documents (prefix HD). These standards are in turn implemented as national standards by all CENELEC member countries. ENs are published unaltered whereas HDs may have minor modifications appropriate to national requirements. Any existing conflicting national standards must be withdrawn when ENs and HDs are implemented by member countries.

Important national and international standards covering LV, MV and HV cables are listed below and some are referred to in appropriate sections of this paper. Standards documents may include clauses related to construction, dimensions, materials and testing.

• BS 5467: LV armoured cables with copper conductors and thermosetting insulation (normally XLPE) for U = 1 kV and U = 3.3 kV. These are cables for non-utility industrial applications.
• BS 6622: MV armoured cables with thermosetting insulation (XLPE and EPR) (see also IEC 60502-2). These are cables for non-utility industrial applications.
• BS 6724: Same scope as BS 5467 but for cables with low emission of smoke and corrosive gases when affected by fire.
• BS 7835: Same scope as BS 6622 but for cables having low emission of smoke and corrosive gases.
• BS 7870-4.10 and -4.11: MV polymeric insulated cables (XLPE and EPR) for use by generation and distribution utilities (implementing applicable parts of HD 620). These are intended as utility cables but may also be specified for industrial applications.
• BS 7912: Cables with XLPE insulation and metal sheath and their accessories for rated voltages from 66 kV ( U m = 72.5 kV) to 132 kV ( U m = 145 kV) (implementation of HD 632). See also IEC 60840.
• BS EN 60228: Conductors, copper and aluminium, stranded and solid, with applicable size range 0.5 to 2500 mm2.
• BS IEC 60287: Calculation of current rating:
  • Part 1: Calculation of losses.
  • Part 2: Calculation of thermal resistances.
  • Part 3: Operating conditions and economic optimisation of cable size.
• IEC 60502: Cables with extruded insulation and their accessories:
  • Part 1: Cables with rated voltages 1 kV ( U m = 1.2 kV) and 3 kV ( U m = 3.6 kV).
  • Part 2: Cables with rated voltages 6 kV ( U m = 7.2 kV) to 30 kV ( U m = 36 kV).
• IEC 60840: Cables with extruded insulation and their accessories for rated voltages above 30 kV ( U m = 36 kV) up to 150 kV ( U m = 170 kV).

Testing of cables

Most of the vast length of LV, MV and HV power cables are buried in the ground and the network owners will hope that there is no need to uncover them except for reasons such as diversion or renewal. Faults in underground cables are very expensive to repair, especially cables installed under roads and pavements. Repairing a fault will probably involve cutting out a length of cable either side of the fault and installing a new length together with joints to connect to the existing cable.

The wide spectrum of tests applied to cables before they are put into service cannot guarantee long life but can at least seek out design weaknesses and provide checks on material and manufacturing quality. Procedures and requirements for testing cables are written into the appropriate cable standards. Test categories and requirements applicable to polymeric cables are summarised as follows.

Tests are grouped in three categories:

• Routine tests made by the manufacturer on each manufactured batch (normally drum length) to check that it meets the specified requirements.
• Sample tests made by the manufacturer on samples of completed cable or components taken from a cable at a specified frequency to verify that the finished product meets the specified requirements.
• Type tests made on a type of cable or system before supplying on a general commercial basis. They are intended to demonstrate satisfactory performance against requirements of the appropriate standard. Type tests are not normally repeated unless changes are made to materials, manufacturing process or design, which might affect performance characteristics.

In addition, there are usually Tests after installation, intended to demonstrate the integrity of the cable and accessories after installation and before use.

Routine testing of all cable lengths is done on the manufacturing site. For LV cables the procedure involves a conductor resistance test and a voltage test between each conductor and the earth components of the cable. Additional tests on MV and HV cables include a metallic screen resistance test and a partial discharge test.

Sample tests are carried out from time to time on lengths of cable taken from a production batch. For LV cables they are non-electrical tests to verify dimensions and constructional characteristics. MV and HV cables are subjected to more extensive testing of material components and electrical tests including AC voltage withstand and partial discharge.

Type tests are intended as qualification of a new cable design or an existing cable whose dimensions or materials have changed to an extent that might affect performance. In contrast to routine and sample tests, type tests are not intended to be repeated and there is no expiry date for the validity of the test report. This is fortunate because type testing is an expensive and lengthy procedure with MV and HV cables, especially if the job is done by one of the well-known independent testing organisations.

Type tests on LV cables are mostly non-electrical and similar to routine and sample tests. Type tests on MV and HV cables and systems are more searching. In addition to a multiplicity of non-electrical tests, the following electrical tests are found in the appropriate standards, though not necessarily all or in the same sequence. Most tests are carried out on the same cable sample with minimum length 10 m. The following summary refers to MV/HV cables up to 132 kV ( U m = 145 kV).

Bending test

Cables for fixed installations are not designed for repeated flexing but must undergo bending during manufacture, drumming and installation without damage. The bending test involves three reverse bends around a specified diameter, followed by a partial discharge test.

Partial discharge test

Partial discharges are low energy breakdowns occurring in gas-filled voids within the insulation body or at interfaces between insulation and conductive screens. Partial discharges within a cable at working voltage are a likely cause of full breakdown at some indeterminate time during the life of the cable. Measurements of partial discharge inception voltage are made at a specified test voltage above Uo using specialised equipment. The individual discharge measurement unit is pico-coulombs (pC) and typical requirements in the standards are ≤10 pC at 2 U o or ≤5 pC at 1.5 U o. The expectation with modern cables is zero partial discharge level but background electrical noise in test locations such as cable factories makes it difficult or impossible to achieve measurement sensitivities less than a few pC. Skilled operators of this test should, however, be able to distinguish partial discharges in the cable from background noise.

Measurement of loss factor ( tan δ)

Dielectric losses are low in polymeric insulation compared with impregnated paper but since there is some energy loss, limits are prescribed through a range of test voltages and conductor temperature. For 66 and 132 kV XLPE cables, BS 7912 and IEC 60840 specify a maximum loss factor of 1 × 10 −3. For MV cables higher loss factors are acceptable, with the maximum value specified in BS 6622 for XLPE insulation being 4 × 10 −3 at U o. EPR is a higher loss material which makes it less favoured for higher voltage cable insulation. For MV applications the limiting tanδ value for EPR is 20 × 10 −3.

Heating cycle voltage test

Heating cycle tests simulate the daily variation of load demand characteristic of utility networks. The well-established standard heating cycle lasts 8 h. Current is induced in the test cable in order to bring the conductor up to 5 K above maximum temperature, and this is held for 2 h, followed by a cooling period to near ambient temperature. This cycling imposes repeated thermo-mechanical stresses on the cable components. During the test the cable is energised at 2 U o. The full test is normally 20 cycles with periodic measurement of partial discharge inception voltage.

Lightning impulse voltage test

All type test regimes for cables with U m = 3.6 kV and above include a test to simulate lightning strikes. The transient surge voltage resulting from a lightning strike will be attenuated along the cable and so the more vulnerable section of cable will be close to its termination to an overhead line. Impulse generators in test laboratories are set up to deliver a single steep-front wave, normally with a front time of 1.2 μs and a tail of 50 μs to 50% voltage decay. The test cable is subjected to 10 positive and 10 negative shots at the prescribed peak voltage. In UK standards examples of impulse withstand peak voltages are 95 kV for 11 kV cables, 194 kV for 33 kV cables and 650 kV for 132 kV cables. The impulse test is normally performed with the cable at maximum working temperature.

HV test

In this test an AC voltage typically 3 U o or 4 U o is applied to the test sample for a period of 4 h. For paper cables this was a relatively searching test but no modern polymeric cable should suffer any risk of failure, given the short length of test cable and the relatively short time under stress.

Tests after installation

Given the practical difficulties of bringing large or complex test equipment to installation sites, voltage tests after installation can do little more than identify major faults associated with the installation, and these are more likely to be associated with joints and terminations rather than the cable itself. Test voltage sources are normally DC or very low frequency. Test voltage is applied between phase conductors and metallic earth components of the cable. In addition, integrity of polymeric oversheaths may be checked by applying a modest test voltage between the metallic earth components of the cable (disconnected from earth) and the surrounding earth.