Industrial Power Transformers -- Operation and maintenance [part 2]

Home | Articles | Forum | Glossary | Books


The subject of neutral grounding is a complex one and, whenever it is discussed by electrical engineers, views are varied and the discussion lengthy. It can and has been made the subject of entire textbooks, so that in devoting no more than part of a section to the topic it is only possible to briefly look at the principal aspects in so far as they affect transformer design and operation. Practices vary in different countries, and even within different utilities in the same country. From time to time over the years individual utilities have had occasion to re-examine their practices and this has sometimes resulted in detail changes being made to them. Fortunately for transformer designers, grounding of a sys tem neutral can only fall into one of three categories. These are:

(1) Neutral solidly grounded (2) Neutral grounded via an impedance (3) Neutral isolated and because of the problems and disadvantages of the third alternative, it is unlikely that it will be encountered in practice so that it is only necessary to be able to design for the first two.

It is intended mainly in this section to examine grounding practices in the UK, where the guiding principles in relation to grounding are determined by statute, in the form of The Electricity Supply Regulations 1988.

The above regulations replaced those of 1937 and the Electricity (Overhead Lines) Regulations 1970 as well as certain sections of the Schedule to the Electric Lighting (Clauses) Act 1899, and they represent mainly a rationalization and updating process rather than any major change of UK practice. Part II of the 1988 regulations contains the provisions relating to grounding. It says that:

• Every electrical system rated at greater than 50 V shall be connected to ground.

• How that ground connection is to be made differs between HV and LV systems.

Low voltage is defined as exceeding 50 V but not exceeding 1000 V and is mainly referring 415 V distribution networks. In the case of an HV system, beyond the requirement that it shall be connected to ground, the method of making the connection is not specified, but for an LV system the regulations say that 'no impedance shall be inserted in any connection with ground … other than that required for the operation of switching devices, instruments, control or telemetering equipment.' In other words LV systems must be solidly grounded. The system of protective multiple grounding, which can be advantageous on 415 V distribution networks in some situations, is permitted on LVs systems subject to certain other conditions but this still requires that the neutral should be solidly grounded 'at or as near as is reasonably practicable to the source of voltage.'

Grounding of high-voltage systems

As stated above, the statutory requirement in the UK is that basically all electrical systems should be connected to ground, so a discussion of the technical merits and demerits is somewhat academic. However, it is essential that readers of a volume such as this understand these fully, so they may be set out as follows.

Advantages of connecting a high-voltage system to ground:

• A ground fault effectively becomes a short circuit from line to neutral. The high-voltage oscillations to which systems having isolated neutrals are susceptible and which can cause serious damage to such systems, are reduced to a minimum, and consequently the factor of safety of the system against ground faults is largely increased. This reasoning applies to systems having overhead lines or underground cables, though to a greater extent the former.

• A grounded neutral allows rapid operation of protection immediately a ground fault occurs on the system. In HV networks most of the line faults take place to ground. Particularly in the case of underground cables, were these on a system employing an isolated neutral, these would take the form of a site of intense arcing activity, which in the case of multicore cables, would result ultimately in a short circuit between phases. The grounded neutral in conjunction with sensitive ground fault protection, results in the faulty section being isolated at an early stage of the fault.

• If the neutral is solidly grounded, the voltage of any live conductor cannot exceed the voltage from line to neutral. As under such conditions the neutral point will be at zero potential, it is possible to effect appreciable reductions in the insulation to ground of cables and overhead lines, which produces a corresponding saving in cost. It is also possible to make similar insulation reductions in transformers and, by the use of non-uniform insulation, make further reductions in the amount of insulation applied to the neutral end of HV windings. In the UK, non-uniform insulation is used for system volt ages of 132 kV and above.

A stable ground fault on one line of a system having an isolated neutral raises the voltage of the two sound lines to full line voltage above ground, which is maintained so long as the fault persists. The insulation of all equipment connected to the sound lines is subjected to this higher voltage, and although it may be able to withstand some overvoltage, it will eventually fail. In extra high-voltage systems, because of capacitance effects, the voltage of the two sound lines may, initially, reach a value approaching twice the normal line voltage by the same phenomenon as that of voltage doubling which takes place when switching a pure capacitance into circuit, and the insulation of the system will be correspondingly overstressed.

• On an ungrounded system the voltage to ground of any line conductor may have any value up to the breakdown value of the insulation to ground, even though the normal voltage between lines and from line to neutral is maintained.

Such a condition may easily arise from capacitance effects on systems having overhead lines, as these are particularly subject to induced static charge from adjacent charged clouds, dust, sleet, fog and rain, and to changes in altitude of the lines. If provision is not made for limiting these induced charges, gradual accumulation takes place, and the line and the equipment connected to it may reach a high 'floating' potential above ground until this is relieved by breakdown to ground of the line or machine insulation or by the operation of co-ordinating gaps or surge arresters.

If, however, the neutral point is grounded either directly or through a cur rent limiting device, the induced static charges are conducted to ground as they appear, and all danger to the insulation of the line and equipment is removed. No part of a solidly grounded neutral system can reach a voltage above ground greater than the normal voltage from line to neutral.

Disadvantages of connecting a high-voltage system to ground

• The only disadvantage of connecting an HV system to ground is that this introduces the first ground from the outset and it thus increases the susceptibility to ground faults. This can be inconvenient in the case of a long overhead line, particularly in areas of high lightning incidence, however, such faults are usually of a transient nature and normally cleared immediately the line is tripped so that delayed auto-reclosure of the line circuit quickly restores supplies.

It is clear, therefore, that the advantages of connection to ground far outweigh the disadvantages. For transformer designers by far the most significant advantage is the ability to utilize non-uniform insulation.

Multiple grounding

One notable difference between the Electricity Supply Regulations of 1988 and those which preceded them, is the attitude to multiple grounding. The regulations of 1937 required that each system should be grounded at one point only and stated that interconnection of systems which were each grounded at one point was not permitted except by special permission of the Electricity Commissioners with the concurrence of the Postmaster-General, who at that time had statutory responsibility for telecommunications. The reason for this was, of course, concern that grounding a system at more than one point would lead to the circulation of harmonic currents via the multiple ground points. As explained in Section 2, the third-order harmonic voltages of a three-phase sys tem are in phase with each other so that if two points of the system are grounded concurrently, the third-order harmonic voltages will act to produce circulating currents. The higher-frequency components, in particular, of these circulating currents can cause interference with telecommunications circuits and this was the cause of the concern to the Postmaster-General. Although the current regulations have removed the statutory limitation on grounding a system at more than one point, the requirement that the supply system must not cause interference with telecommunications equipment is covered by the more general provisions of the European Union's Directive concerning electromagnetic compatibility which places the onus on all users of electrical equipment to ensure that it does not cause electromagnetic interference. How this is achieved is the responsibility of the user of the equipment and there could be sound technical reasons for wishing to have more than one ground on the system. In this situation the user may elect to guard against generating interference by the use of a third harmonic suppresser, that is, a device, usually a reactor, in one of the neutral connections, which has minimal impedance to 50 or 60 Hz currents but much higher impedance to higher-order harmonics.

Solid versus impedance grounding of transformer neutral points

As indicated above, for HV systems, the Regulations are not specific as to how the system grounding should be carried out. From a practical viewpoint however, if it is required to utilize non-uniform insulation, it is necessary to ensure that the voltage of the neutral remains at the lowest practicable level for all fault conditions, that is, a solid ground connection is required. The economic benefits of non-uniform insulation become marked at 132 kV and above and it is thus standard practice throughout the UK to solidly ground systems of 132 kV and above. The option for impedance grounding is thus available without any economic penalty as far as the transformer insulation is concerned for all other systems classed as HV systems. This in practical terms means systems from 66 kV down to 3.3 kV inclusive.

The next decision to be made is whether impedance grounding will be beneficial if utilized for these systems and, if so, what criteria should be used to decide the value and type of impedance. In answering this question it is necessary to consider why impedance grounding might be desirable, and the reason for this is that it limits the current which will flow in the event of a ground fault.

Hence, the damage caused at the point of the fault is greatly reduced. Applying this logic alone would result in the option for a high value of impedance, but the problem then is that some ground faults can themselves have a high impedance and in this situation there could be a problem that the protection will be slow in detecting their existence. Usually the level of impedance selected is such as to result in the flow of system full-load line current for a solid, that is zero impedance, ground fault. On this basis a 60 MVA transformer providing a 33 kV supply to a grid bulk supply point would have the 33 kV neutral grounded with a value of impedance to limit the ground fault current to

60 000 000 3 33 000 1050


_ A.

It was the practice of the UK Electricity Supply Industry to place a lower limit on the value of ground fault current, so that for a 30 MVA, 33 kV transformer supply the impedance would be such as to allow a fault current of 750 A rather than 525 A. [1] Other supply companies may wish to standardize on, say, 1000 A as a convenient round figure.

[1. The exception to this rule was at CEGB generating stations from the mid-1970s at which the generator ground fault current was limited to the very low value of about 10 A. ]

Grounding of delta-connected transformers

In the above example it is likely that the transformer providing the 33 kV bulk supply would have its primary connected at 132 kV, which, to take advantage of the use of non-uniform insulation, would have its HV winding star connected with the neutral solidly grounded. The 33 kV winding would thus probably be connected in delta and hence would not provide a 33 kV system neutral point for connection to ground. Hence a neutral point must be provided artificially by the use of auxiliary apparatus specially designed for the purpose.

This usually takes the form of an interconnected star neutral grounding transformer, although very occasionally a star delta transformer might be used.

FIG. 5 Interconnected star neutral grounding transformer

FIG. 6 Three-phase, star/delta neutral grounding transformer

The two schemes are shown diagrammatically in Figs 6.5 and 6.6. The interconnected star connection is described in Section 2. It is effectively a one to-one autotransformer with the windings so arranged that, while the voltages from each line to ground are maintained under normal operating conditions, a minimum impedance is offered to the flow of single-phase fault current, such as is produced by a ground fault on one line of a system having a grounded neutral. Under normal operating conditions the currents flowing through the windings are the magnetizing currents of the grounding transformer only, but the windings are designed to carry the maximum possible fault current to which they may be subjected, usually for a period of 30 seconds. The apparatus is built exactly as a three-phase core type transformer, and is oil immersed.

While the interconnected star grounding transformer is the type most often used for providing an artificial neutral point, an alternative may be adopted in the form of an ordinary three-phase core type transformer having star connected primary windings, the neutral of which is grounded and the line ends connected to the three-phase lines, while the secondary windings are connected in closed delta, but otherwise isolated. Normally, the current taken by the transformer is the magnetizing current only, but under fault conditions the closed delta windings act to distribute the fault currents in all three-phases on the primary side of the transformer, and as primary and secondary fault ampere-turns balance each other, the unit offers a low impedance to the cur rent flow. The transformer is rated on the same basis as outlined for the interconnected star grounding transformer and it is constructed exactly the same as an ordinary power transformer.

For the purpose of fault current limitation, resistors may be used in conjunction with either of the above types of grounding transformer, and they may be inserted between the neutral point and ground, or between the terminals of the grounding transformer and the lines. In the former case one resistor is required, but it must be designed to carry the total fault current, while it should be insulated for a voltage equal to the phase voltage of the system. On the other hand, the neutral point of the grounding transformer windings will rise to a voltage above ground under fault conditions equal to the voltage drop across the grounding resistor, and the transformer windings will have to be insulated for the full line voltage above ground.

While in any case this latter procedure may be adopted, it is not desirable to subject the grounding transformer windings to sudden voltage surges any higher than can be avoided, as the insulated windings are the most vulnerable part of the equipment. If, suitably proportioned resistors are placed between the terminals of the grounding transformer and the lines instead of between neutral and ground, exactly the same purpose is served so far as fault current limitation is concerned, while, the neutral point of the grounding transformer always remains at ground potential, and the windings are not subjected to any high voltages.

On the other hand, the insulators must now be insulated for full line voltage, but this is a relatively easy and cheap procedure. For the same fault current and voltage drop across the resistors the ohmic value of each of those placed between the grounding transformer terminals and lines is 3 times the ohmic value of the single resistor connected between the neutral and the ground, but the current rating of each resistor in the line is one-third of the current rating of a resistor in the neutral, as under fault conditions the three resistors in the lines operate in parallel to give the desired protection.

Value of grounding impedance

For any of the arrangements described above, the magnitude of resistor required can be determined by a simple application of Ohms law:

Neutral grounding apparatus

The most common device used for connection in the HV neutral is the liquid neutral grounding resistor or LNER. These are relatively inexpensive, sturdy and can easily be constructed to carry ground fault currents of the order of up to 1500 A. They are generally designed to carry the fault current for up to 30 seconds. The ohmic value of the resistor is a function of the system volt age to ground and of the permissible fault current. A minor disadvantage of liquid resistors is that they require maintenance in the form of ensuring that the electrolyte is kept topped up and at the correct strength, which might present a slightly increased burden in hot climates and in temperate climates they require heaters to prevent freezing in winter. For this reason metallic resistors are sometimes preferred. These may take the form of pressed grids or stainless steel wound modules which can be connected with the appropriate numbers in series and parallel to provide the required voltage and current rating. These have high reliability and ruggedness, their only disadvantage being cost.

An alternative to resistance grounding is the use of an arc suppression coil. The arc suppression coil was first devised by W. Petersen in 1916, and is hence the generally known as a Petersen coil. Use of an arc suppression coil enables a power system to benefit from the advantage normally associated with ungrounded systems without suffering their disadvantages. Basically, it is a reactor connected between the neutral of the supply transformer and ground. The reactance of the coil is tuned to match the capacitance of the power system it is protecting.

As indicated above, the majority of faults on an HV network are ground faults and most of these involve single phase to ground contact of an arcing nature. With an arc suppression coil installed, intermittent faults are made self clearing. This is due to the resonance established between the capacitance of the system and the inductance of the arc suppression coil which results in balancing of the leading and lagging components of current at the point of the fault. Any small residual ground current sufficient to sustain the arc is substantially in phase with the voltage of the faulty conductor, and since both pass through zero at the same instant, the arc is extinguished. The resonance delays the recovery voltage build-up after arc extinction which enables the dielectric strength of the insulation at the point of the fault to recover and prevent restriking of the arc. FIG. 7 shows a typical oscillogram of recovery volt age following arc extinction in such an installation.

FIG. 7 Recovery voltage after the initial arc extinction

FIG. 8 Voltages and currents at the point of a ground fault on a Petersen coil grounded system.

In the event of a sustained phase to ground fault, the arc suppression coil allows the power system to be operated in a faulted condition until the fault can be located and removed. The residual fault current is normally of the order of 5-10 percent of the total capacitative fault current. The phasor relationship between the voltages on the three-phase conductors and the currents through the fault and the arc suppression coil is shown in FIG. 8. Nowadays, solid-state control devices can be used in conjunction with arc suppression coils which, in conjunction with automatic switching of taps on the arc suppression coil during the fault, enable optimum compensation to be achieved. This technique is particularly useful for systems with multiple feeders, where a ground fault on one feeder results in a different magnitude of fault current to a ground fault on another feeder.

The insulation level of all the plant and apparatus on the system on which arc suppression coils are installed must be adequate to allow operation for a period with one line grounded, and it is generally found uneconomic to install them on systems operating above 66 kV. Up to this voltage, the standard insulation level, without grading, is likely to be employed for all transformers. It is recommended that a higher insulation level should be considered if operation of the system with one line grounded is likely for more than 8 hours in any 24, or more than 125 hours in any year.

The choice of whether to continue operation with a sustained fault on the network lies with the operator. Although it has been shown that arc suppression coils allow this, other factors must be considered, the most important being the safety of personnel. For example, the fault may have been caused by a broken line conductor which would clearly constitute a danger. Should the utility decide not to operate with sustained faults the faulted section must be isolated as soon as a sustained fault is detected. Previously it was common practice to short circuit the arc suppression coil after a specified time to allow protection relays to operate. When the coil is short circuited a significant in-rush of fault current may occur, which would cause a voltage dip. Now, using modern protection devices, it is possible to leave the arc suppression coil in service. Isolation of the faulted section can be carried out, for example, using admittance sensing relays which can determine changes in the admittances of the lines, instead of overcurrent relays as traditionally used.

Ground connection

When dealing with the question of neutral point grounding it is important to give careful attention to the ground connection itself, that is, to the electrode buried in the ground for the purpose of obtaining a sound ground. If the grounding system is not carefully installed and maintained, then serious danger may occur under fault conditions from touch and step potentials (see below).

For obtaining a direct ground contact copper or cast iron plates, iron pipes, driven copper rods, copper strips or galvanized iron strips may be employed. It is not always appreciated that it is very difficult to obtain resistance values of less than about 2 Ohm from a single ground plate, and often it is still more difficult to maintain the value after the grounding system has been installed for some time.

On account of this it is usual to install several ground plates, pipes, etc., in parallel, so that the combined resistance of the installation is reduced to a reasonably low value of 1 Ohm or less. Where a parallel arrangement is employed, each plate, rod, etc., should be installed outside the resistance area of any other. Strictly, this requires a separation of the order of 10m which, however, can often be reduced without increasing the total resistance by more than a few percent.

The chief points to be borne in mind when installing a grounding equipment are, that it must possess sufficient total cross-sectional area to carry the maximum fault current, and it must have a very low resistance in order to keep down to a safe value the potential gradient in the ground surrounding the plates, etc., under fault conditions. As most of the resistance of the grounding system exists in the immediate vicinity of the plates, etc., the potential gradient in the ground under fault conditions is naturally similarly located, and in order that this shall be kept to such a value as will not endanger life, the current density in the ground installation should be kept to a low figure either by using a number of the plates, pipes, etc., in parallel, or else by burying to a considerable depth, making the connection to them by means of insulated cable. The former arrangement is one which can best be adopted where there are facilities for obtaining good grounds, but in cases where, on account of the nature of the ground, it has been difficult to obtain a good ground, driven rods have been sunk to a depth of 10 m and more. The maximum current density around an electrode is, in general, minimized by making its dimensions in one direction large with respect to those in the other two, as is the case with a pipe, rod or strip.

Ground plates are usually made of galvanized cast iron not less than 12 mm thick, or of copper not less than 2.5 mm in thickness, the sizes in common use being between 0.6 and 1.2 m^2. If a ground of greater conductivity is required, it is preferable to use two or more such plates in parallel. Ground pipes may be of cast iron up to 100 mm diameter, 12 mm thick and 2.5-3 m long, and they must be buried in a similar manner to ground plates. Alternatively, in small installations, driven mild steel pipes of 30-50 mm diameter are some times employed.

Where the driving technique is adopted, copper rods are more generally used. These consist of 12-20 mm diameter copper in sections of 1-1.5 m, with screwed couplers and a driving tip. Deeply driven rods are effective where the soil resistivity decreases with depth but, in general, a group of shorter rods arranged in parallel is to be preferred.

In cases where high-resistivity soil (or impenetrable strata) underlies a shallow surface layer of low-resistivity soil a grounding installation may be made up of untinned copper strip of section not less than 20 by 3 mm or of bare stranded copper conductor.

If a site can be utilized which is naturally moist and poorly drained, it is likely to exhibit a low soil resistivity. A site kept moist by running water should, however, be avoided. The conductivity of a site may be improved by chemical treatment of the soil, but it should be verified that there will be no deleterious effect on the electrode material. To ensure maximum conductivity, ground electrodes must be in firm direct contact with the ground.

It is most important that the connections from the neutral or auxiliary apparatus to the ground installation itself should be of ample cross-sectional area, so that there is adequate margin over the maximum fault current, and so that no abnormal voltage drop occurs over their length; the connections to the grounding structure having ample surface contact.

Grounding of low-voltage systems

As indicated in the introduction to this section, low-voltage systems are defined in the UK as being above 50 V but below 1000 V and this is mainly intended to embrace all industrial three-phase systems operating at 400 V and domestic single-phase 230 V systems supplied from one phase and neutral of the 400 V network. Although the recent development of the ground leakage circuit breaker has resulted in some changes to safety philosophy, these systems are still mainly protected by fuses, and in order to provide maximum protection to personnel by ensuring rapid fuse operation and disconnection of faulty equipment, the systems are designed to have the lowest practicable ground loop impedance.

This means that a solid neutral ground connection must be provided.

The fundamental importance of the solid ground connection is underlined by its embodiment in the 1988 Supply Regulations and also the benefits of the sys tem of protective multiple grounding in assisting the achievement of low ground loop impedance in areas where this might not otherwise be possible is acknowledged by the inclusion of a clause setting down how this is to be carried out.

The requirement for solid grounding of the LV neutral also aims to ensure that the likelihood of the presence of any voltage above normal appearing in the LV circuit is reduced to a minimum since, in the event of insulation breakdown between HV and LV windings of the step-down transformer the resulting ground fault on the HV system should ensure rapid operation of the HV system ground fault protection. The exception is when the HV side of the transformer is connected to ground through a continuously rated arc suppression coil. In this case the point of fault between windings remains at close to its potential determined by its location in the LV winding, that is the voltages on the LV system change very little from those occurring under healthy conditions, and the distribution of voltages on the HV side is adjusted accordingly. In practice, breakdown between HV and LV windings of any transformer connected to an HV system is such an unlikely occurrence as to be discounted in the carrying out of any risk assessment.

Grounding system design

At the start of this section the view was expressed that the subject of neutral grounding was a complex one, so that, clearly, the design of grounding systems is not a topic to be covered in a few paragraphs in a textbook dealing with transformers. However, it is necessary to say a little about the subject of grounding system design, at least to explain the philosophy, which has changed some what in recent years and, in particular, since earlier editions of this work were written. The most significant change is that now the grounding system must be designed to ensure that the potentials in its vicinity during a fault are below appropriate limits. Previously it was established practice to design the grounding system to achieve a certain impedance value.

When a ground fault occurs and current flows to ground via a ground electrode, or system of electrodes, the potential on the electrodes or any equipment connected to them will rise above true ground potential. This potential rise can be particularly substantial, of the order of several thousand volts in the case of large substations subjected to severe faults. The objective in seeking to obtain a satisfactory grounding system design is to ensure 'safety to personnel' by avoiding the creation of dangerous touch, step or transferred potentials, whilst acknowledging that the ground potential rise under severe fault conditions must inevitably exist.

The philosophy will be made clearer by definition of the above terms.

Interpretation of the definitions will be made clear by reference to FIG. 9.

FIG. 9 Differences in ground potential.

When the potential rise of a ground electrode occurs due to a fault, this will form a potential gradient in the surrounding ground. For a single electrode the potential gradient will be as shown in the figure. A person in the vicinity of this electrode may be subjected to three different types of hazard as a result of this potential gradient:

• Step potential: person 'a' in the figure illustrates 'step potential.' Here the potential difference V1 seen by the body is limited to the value between two points on the ground separated by the distance of one pace. Since the potential gradient in the ground is greatest immediately adjacent to the electrode area, it follows that the maximum step potential under ground fault conditions will be experienced by a person who has one foot in the area of maximum rise and one foot one step towards true ground.

• Touch potential: person 'b' in the figure illustrates 'touch potential.' Here the potential difference V2 seen by the body is the result of hand-to-both feet contact. Again the highest potential will occur if there were a metal structure on the edge of the highest-potential area, and the person stood one pace away and touched the metal. The risk from this type of contact is higher than for step potential because the voltage is applied across the body and could affect the heart muscles.

• Transferred potential: the distance between the high-potential area and that of true ground may be sufficient to form a physical separation rendering a person in the high-potential area immune from the possibility of simultaneous contact with zero potential. However, a metal object having sufficient length, such as a fence, cable sheath or cable core may be located in a manner that would bridge this physical separation. By such means, zero ground potential may be transferred into a high-potential area or vice versa. Person 'c' in FIG. 9 illustrates the case of a high potential being transferred into a zero potential area via the amour of a cable. If the amour is bonded to ground at the substation, that is the fault location, the voltage V3 will be the full 'rise of ground potential of the substation.' In the case illustrated the person at 'c' is making simultaneous contact hand to hand with the cable sheath and true ground. However, if the person is standing on true ground then the voltage V3 seen by the body could be hand-to-both-feet contact. Person 'd' represents the case of zero potential being transferred to a high-potential area via a cable core which is grounded at the remote point. In this case, the voltage V4 is lower than V3 which represents the substation rise of ground potential, because person 'd' is located some distance from the main ground electrode and therefore benefits from the ground potential gradient. Clearly, if person 'd' had been on or touching the main electrode he would have experienced the full rise of ground potential V3.

It will be apparent from the above that transferred potentials can present the greatest risk, since the shock voltage can be equal to the full rise of ground potential and not a fraction of it as is the case with step or touch potentials.

Historically limits on transfer potentials have been set at 650 and 430V in the UK, depending on the type of installation, above which special precautions are required. The higher value is normally taken to apply for high reliability systems having high-speed protection. No limiting clearance time is quoted for these systems but it is generally accepted that these will clear in 0.2 seconds. The lower figure is for systems protected by overcurrent protection, and although again no limiting clearance time is specified, a time of 0.46 seconds is generally assumed.

If the ground electrode system cannot be designed to comply with the above criteria, then the type of special precautions which might be considered to protect against transferred potentials is the provision of local bonding to ensure that all metalwork to which simultaneous contact can be made is at the same potential. Consideration might also be given to restricting telephone and Supervisory Control and Data Acquisition (SCADA) connections with remote locations to those using fiber optic cables. Guard rings buried at increasing depths around an electrode can be used to modify the ground surface potential to protect against step potentials.

For those contemplating the design of a grounding system a number of standards and codes of practice are available. In the UK, the most important of these are:

• BS 7354: 1990 'Code of practice for design of high-voltage open terminal stations.'

• BS 7430: 1998 'Code of practice for grounding.'

• BS 7671: 2001 'Requirements for electrical installations. IEE wiring regulations. Sixteenth edition.'

• EA Engineering Recommendation S34: 1986 'A guide for assessing the rise of ground potential at substation sites.'

• EA Technical Specification 41-24: 1994 'Guidelines for the design, testing and maintenance of main grounding systems in substations.' The book 'Grounding Practice' published by the Copper Development Association, also contains much useful information.

Top of Page

PREV.   NEXT   Guide Index HOME