1 Overload currents

These are currents higher than those intended to be present in the system. If such currents
persist they will result in an increase in conductor temperature, and hence a rise in
insulation temperature. High conductor temperatures are of little consequence except that
the resistance of the conductor will be increased leading to greater levels of voltage drop.

Insulation cannot tolerate high temperatures since they will lead to deterioration and
eventually failure. The most common insulation material is p.v.c. If it becomes too hot it
softens, allowing conductors which press against it (and this will happen in all cases
where a conductor is bent) to migrate through it so that they come close to, or even move
beyond, the insulation surface. For this reason, p.v.c. insulation should not normally run
at temperatures higher than 70°C, whereas under overload conditions it may have
allowable temperatures up to 115°C for a short period during transient conditions.

2 Short circuit currents

These currents will only occur under fault conditions, and may be very high indeed. As
we shall shortly show (see {3.6.3 and 3.6.4}) such currents will open the protective devices
very quickly. These currents will not flow for long periods, so that under such short-term
circumstances the temperature of p.v.c. insulation may be allowed to rise to 160°C.

The clearance time of the protective device is governed by the adiabatic equation which
is considered more fully in {3.7.3}.

2.7 - Undervoltage

This chapter of the Regulations deals with the prevention of dangers that could occur if voltage
falls to a level too low for safe operation of plant and protective devices. Another problem covered
is the danger that may arise when voltage is suddenly restored to a system which has previously
been on a lower voltage or without voltage at all. For example, a machine which has stopped due
to voltage falling to a low level may be dangerous if it restarts suddenly and unexpectedly when
full voltage returns. A motor starter with built-in undervoltage protection will be explained in
{7.15.1}.

The attention of the installer and the designer is drawn to the possibility that low voltage may
cause equipment damage. Should such damage occur, it must not cause danger. Where
equipment is capable of operating safely at low voltage for a short time, a time delay may be used
to prevent switching off at once when under-voltage occurs. This system may prevent plant
stoppages due to very short time voltage failures. However, such a delay must not prevent the
immediate operation of protective systems.

3.2.1 - Switch positions

A switch is defined as a device which is capable of making or breaking a circuit under

normal and under overload conditions. It can make, but will not necessarily break, a short
circuit, which should be broken by the overload protecting fuse or circuit breaker. A
switching device may be marked with ON and OFF positions, or increasingly, the
numbers 1 for ON and 0 for OFF are being used.
A semiconductor device is often used for switching some lighting and heating circuits,
but will not be suitable for disconnecting overloads; thus, it must be backed up by a
mechanical switch. The semiconductor is a functional switch but must NOT be used as an
isolator.

{Figure 3.1} shows which poles of the supply need to be broken by the controlling
switches. For the TN-S system (earth terminal provided by the Electricity Company), the
TNC-S system (protective multiple earthing) and the TT system (no earth provided at the
supply), all phase conductors MUST be switched, but NOT the protective (earth)
conductor.

The neutral conductor need not be broken except for:

1 - the main switch in a single-phase installation, or
2 - heating appliances where the element can be touched, or
3 - autotransformers (not exceeding 1.5 kV) feeding discharge lamps

The neutral will need to he disconnected for periodic testing, and provision must be made
for this; it is important that the means of disconnection is accessible and can only be
completed with the use of a tool.

The protective conductor should never be switched, except when the supply can he taken
from either of two sources with earth Systems which must slot be connected together. In
this case the switches needed in the protective conductors must be linked to the phase
switches so that it is impossible for the supply to be provided unless the earthing
connection is present.

Fig 3.1 Supply system broken by switches

(a)TN-C Systems (b) TN-S, TN-C-S (c) TT Systems

Every circuit must be provided with a switching system so that it can be interrupted on
load. In practice, this does not mean a switch controlling each separate circuit; provided
that loads are controlled by switches, a number of circuits may be under the overall
control of one main switch. An example is the consumer unit used in the typical house,
where there is usually only one main switch to control all the circuits, which are provided
with individual switches to operate separate lights, heaters, and so on. If an installation is
supplied from more than one source there must he a separate main switch for each source,
and each must be clearly marked to warn the person switching off the supplies that more
than one switch needs to he operated.

It should he noted that a residual current device (RCD) may be used as a switch provided
that its rated breaking capacity is high enough.

3.6.4 -Circuit breakers

Circuit breakers operate using one or both of two principles. They are:

1. - Thermal operation relies on the extra heat produced by the high current warming a
bimetal strip, which bends to trip the operating contacts,

2. - Magnetic operation is due to the magnetic field set up by a coil carrying the current,
which attracts an iron part to trip the breaker when the current becomes large enough.

Thermal operation is slow, so it is not suitable for the speedy disconnection required to
clear fault currents. However, it is ideal for operation in the event of small but prolonged
overload currents. Magnetic operation can be very fast and so it is used for breaking fault
currents; in many cases, both thermal and magnetic operation are combined to make the
circuit breaker more suitable for both overload and fault protection. It must be
remembered that the mechanical operation of opening the contacts takes a definite
minimum time, typically 20 ms, so there can never be the possibility of truly
instantaneous operation. A typical time/current characteristic for a circuit breaker is
shown {Fig. 3.12(b)}

All circuit breakers must have an indication of their current rating. Miniature circuit
breakers have fixed ratings but moulded case types can be adjusted. Such adjustment
must require the use of a key or a tool so that the rating is unlikely to be altered except by
a skilled or instructed person.

There are many types and ratings of moulded case circuit breakers, and if they are used,
reference should be made to supplier's literature for their characteristics. Miniature circuit
breakers are manufactured in fixed ratings from 5 A to 100 A for some types, and in six
types, type B giving the closest protection. Operating characteristics for some of the more
commonly used ratings of types 1, 3, B and D are shown in {Figs 3.16 to 3.19}. The
characteristics of Type C circuit breakers are very similar to those of Type 3.
BS3871, which specified the miniature circuit breakers Types 1 to 4 was withdrawn in
1994 and has been replaced with BS EN 60898:1991 (EN stands for "European norm"),
although it is possible that circuit breakers to the old standard will still be on sale for five
years from its withdrawal. In due course, it is intended that only types B, C and D will be
available, although it will be many years before the older types cease to be used. Short
circuit ratings for the newer types will be a minimum of 3 kA and may be as high as 25
kA - the older types had short circuit ratings which were rarely higher than 9 kA.

The time/current characteristics of all circuit breakers {Figs 3.16 to 3.19} have a vertical
section where there is a wide range of operating times for a certain current. Hence, with a
fixed supply voltage, the maximum earth fault loop impedance is also fixed over this
range of time. The operating current during the time concerned is a fixed multiple of the
rated current. For example, a Type 2 MCB has a multiple of 7 (from {Table 3.3}) so a 30
A device of this type will operate over the time range of 0.04 s to 8 s at a current of 7 x
30 A = 210 A.

Table 3.3 Operating time ranges and current multiples for MCBs
over fixed current section of characteristic

MCB Type Range of operating times (s) Current multiple of rating

1 0.04 to 20 x4

2 0.04 to 8 x7

3 0.04 to 5 x10

B 0.04 to 13 x5

C 0.04 to 5 x10

D 0.05 to 3 x20

Table 3.4 A comparison of types of protective device

Miniature circuit
Semi-enclosed fuses HBC fuses
breakers

Very low initial cost Medium initial cost High initial cost

Low replacement
Medium replacement cost Zero replacement cost
cost

Low breaking Very high breaking Medium breaking

capacity capacity capacity

Table 3.5 Comparison of miniature circuit breaker types

 Will not
Type Will trip in Typical application
trip in

100ms at 100ms at
rating rating

1 2.7 x 4x Low inrush currents (domestic installations)

2 4x 7x General purpose use

3 7x 10 x High inrush currents (motor circuits)

B 3x 5x General purpose use (close protection)

Commercial and industrial applications with

C 5x 10 x
fluorescent fittings

Applications where high in-rush currents

D 10 x 50 x
are likely (transformers, welding machines)

{Table 3.4} shows a comparison of the three main types of protective device in terms of
cost, whilst {Table 3.5} compares the available types of MCB.

3.7.1 -Introduction

The overload currents considered in the previous section are never likely to be more than
two to three times the normal rated current. Fault currents, on the other hand, can well be
several hundreds, or even several thousands of times normal. In the event of a short
circuit or an earth fault causing such current, the circuit must be broken before the cables
are damaged by high temperatures or by electromechanical stresses. The latter stresses
will be due to the force on a current carrying conductor which is subject to the magnetic
field set up by adjacent conductors. This force is proportional to the current, and to the
magnetic field strength. Since the field strength also depends on the current, force is
proportional to the square of the current. If the current is one thousand times normal,
force will be one million times greater than usual! Fault protection must not only be able
to break such currents, but to do so before damage results. Abrasion of cable insulation
by movement is usually prevented by normal fixings or by being enclosed in conduit or
trunking. Support must he provided to cables in busbar chambers.

3.7.2 -Prospective short-circuit current (PSC)

The current which is likely to flow in a circuit if line and neutral cables are short circuited
is called the prospective short circuit current (PSC). It is the largest current which can
flow in the system. and protective devices must be capable of breaking it safely. The
breaking capacity of a fuse or of a circuit breaker is one of the factors which need to be
considered in its selection. Consumer units to BS EN 60439-4 and BS 88 (HBC) fuses are
capable of breaking any probable prospective short-circuit current, but before using other
equipment the installer must make sure that their breaking capacity exceeds the PSC at
the point at which they are to he installed.
The effective breaking capacity of overcurrent devices varies widely with their
construction. Semi-enclosed fuses are capable of breaking currents of 1 kA to 4 kA
depending on their type. whilst cartridge fuses to BS 1361 will safely break at 16.5 kA
for type 1 or 33 kA for type II. BS 88 fuses are capable of breaking any possible short-
circuit current. Miniature circuit breakers to BS EN60898 have their rated breaking
capacity marked on their cases in amperes (not kA) although above 10000 A the MCB
may be damaged and lower breaking currents (75% for 10000 A and 50% above that
level) must be used for design purposes.

Prospective short circuit current is driven by the e.m.f. of he secondary winding of the
supply transformer through an impedance made up of the secondary winding and the
cables from the transformer to the fault {Fig 3.21}. The impedance of the cables will
depend on their size and length, so the PSC value will vary throughout the installation,
becoming smaller as the distance from the intake position increases. (313-01-01] requires
the PSC to be 'assessed' by 'calculation, measurement, enquiry or inspection'. In practice,
this can be difficult because it depends to some extent on impedance's which are not only
outside the installation in the supply system, but are also live. If the impedance of the
supply system can be found, a straightforward calculation using the formula of {Fig
3.21} can be used, but this is seldom the case. An alternative is to ask the local Electricity
Company. The problem here is that they are likely to protect themselves by giving a
figure which is usually at least 16 kA in excess of the true value. The problem with using
this figure is that the higher the breaking capacity of fuses and circuit breakers are (and
this must never be less than the PSC for the point at which they are installed), the higher
will be their cost.. {Table 3.6} gives a method of arriving at PSC if the type and length of
the service cable is known,

Fig 3.21 Prospective short circuit current (PSC)

Table 3.6 - Estimation of PSC at the intake position

Length of PSC (kA) up to PSC (kA) over 35mm2

supply cable (m) 25mm2 AI, AI,
 16mm2 Cu supply 25mm2 Cu supply
-
cable cable

5 10.0 12.0

10 7.8 9.3

15 6.0 7.4

20 4.9 6.2

25 4.1 5.3

30 3.5 4.6

40 2.7 3.6

50 2 3.0

The table is not applicable in London, where the density of the distribution system means
that higher values may apply. In this case it will be necessary to consult London
Electricity.

There are two methods for measuring the value of PSC, but these can only be used when
the supply has already been connected. By then, the fuses and circuit breakers will
already be installed.

The first method is to measure the impedance of the supply by determining its voltage
regulation, that is, the amount by which the voltage falls with an increase in current. For
example, consider an installation with a no-load terminal voltage of 240 V. If, when a
current of 40 A flows, the voltage falls to 238 V, the volt drop will be due to the
impedance of the supply.

Thus Zs = systems volt drop = 240 – 238  = 2  = 0.05 

current 40 40
Then PSC = Uo = 240 A = 4800 A or 4.8 kA
Zs 0.05

A second measurement method is to use a loop impedance tester see { 5.3 and 8.6.2 }
connected to phase and neutral (instead of phase and earth) to measure supply impedance.
This can then be used with the supply voltage as above to calculate PSC. Some
manufacturers modify their earth-loop testers so that this connection is made by selecting
'PSC' with a switch. The instrument measures supply voltage, and calculates, then
displays, PSC.

A possible difficulty in measuring PSC, and thus being able to use fuses or circuit
breakers with a lower breaking capacity than that suggested by the Supply Company, is
that the supply may be reinforced. More load may result in extra or different transformers
and cables being installed, which may reduce supply impedance and increase PSC.
 3.7.3 -Operating time

Not only must the short-circuit protection system open the circuit to cut off a fault, but it
must do so quickly enough to prevent both a damaging rise in the conductor insulation
temperature and mechanical damage due to cable movement under the influence of
electro-mechanical force. The time taken for the operation of fuses and circuit breakers of
various types and ratings is shown in {Figs 3.13 to 3.19}. When the prospective short
circuit current (PSC) for the point at which the protection is installed is less than its
breaking capacity there will be no problem.

When a short circuit occurs there will be a high current which must be interrupted
quickly to prevent a rapid rise in conductor temperature.

The position is complicated because the rise in conductor temperature results in an

increase in resistance which leads to an increased loss of energy and increased heating
(W = l²Rt), where W is the energy (J) and R is the resistance . The Regulations make
use of the adiabatic equation which assumes that all the energy dissipated in the
conductor remains within it in the form of heat, because the faulty circuit is opened 50
quickly. The equation is:

t= k²S²
I²

where t = the time for fault current to raise conductor temperature to the highest
permissible level

k = a factor which varies with the type of cable

S = the cross-sectional area of the conductor (mm²)

I = the fault current value (A) - this will be the PSC

Some cable temperatures and values of k for common cables are given in {Table 3.7}.

Table 3.7 Cable temperatures and k values (copper cable)

Assumed initial Limiting final

Insulation material k
temperature (°C) temperature(°C)

p.v.c 70 160 115

85°C p.v.c 85 160 104
90°C thermosetting 90 250 143
Mineral, exposed to touch
70 160 115
or p.v.c. covered
Mineral not exposed to touch 105 250 135
As an example, consider a 10 mm² cable with p.v.c. insulation protected by a 40 A fuse to
BS 88 Part 2 in an installation where the loop impedance between lines at the point where
the fuse is installed is 0.12 . If the supply is 415 V three phase, the prospective short
circuit current (PSC) will be:

I = UL A = 415 A = 3.46 kA
Z 0.12
from {Table 3.7}, k = 115
t = k²S² = 115² x 10² = 0.110s
-I² 34602

{Figure 3.15}shows that a 40 A fuse to BS 88 Part 2 will operate in 0.1 5 when carrying a
current of 400 A. Since the calculated PSC at 3460 A is much greater than 400 A. the
fuse will almost certainly clear the fault in a good deal less than 0.1 s. As this time is less
than that calculated by using the adiabatic equation (0.11 s) the cable will be unharmed in
the event of a short circuit fault.

It is important to appreciate that the adiabatic equation applies to all cables, regardless of
size. Provided that a protective device on the load side of a circuit has a breaking capacity
equal to or larger than the PSC of the circuit then that circuit complies with the PSC
requirements of the Regulations (see {Fig 3.22} and see also the note in {7.15.1} concerning
the use of dual rated fuses for motor protection).

4.1 - Cable insulation materials

Rubber
For many years wiring cables were insulated with vulcanised natural rubber (VIR). Much
cable of this type is still in service, although it is many years since it was last
manufactured. Since the insulation is organic, it is subject to the normal ageing process,
becoming hard and brittle. In this condition it will continue to give satisfactory service
unless it is disturbed, when the rubber cracks and loses its insulating properties. It is
advisable that wiring of this type which is still in service should be replaced by a more
modern cable. Synthetic rubber compounds are used widely for insulation and sheathing
of cables for flexible and for heavy duty applications. Many variations are possible, with
conductor temperature ratings from 60°C to 180°C, as well as resistance to oil, ozone and
ultra-violet radiation depending on the formulation.

Paper
Dry paper is an excellent insulator but loses its insulating properties if it becomes wet.
Dry paper is hygroscopic, that is, it absorbs moisture from the air. It must be sealed to
ensure that there is no contact with the air. Because of this, paper insulated cables are
sheathed with impervious materials, lead being the most common. PILC (paper insulated
lead covered) is traditionally used for heavy power work. The paper insulation is
impregnated with oil or non-draining compound to improve its long-term performance.
Cables of this kind need special jointing methods to ensure that the insulation remains
sealed. This difficulty, as well as the weight of the cable, has led to the widespread use of
p.v.c. and XLPE (thermosetting) insulated cables in place of paper insulated types.
P.V.C.
Polyvinyl chloride (p.v.c.) is now the most usual low voltage cable insulation. It is clean
to handle and is reasonably resistant to oils and other chemicals. When p.v.c. burns, it
emits dense smoke and corrosive hydrogen chloride gas. The physical characteristics of
the material change with temperature: when cold it becomes hard and difficult to strip,
and so BS 7671 specifies that it should not be worked at temperatures below 5°C.
However a special p.v.c. is available which remains flexible at temperatures down to -
20°C.

At high temperatures the material becomes soft so that conductors which are pressing on
the insulation (eg at bends) will 'migrate' through it, sometimes moving to the edge of the
insulation. Because of this property the temperature of general purpose P.V.C. must not
be allowed to exceed 70°C, although versions which will operate safely at temperatures
up to 85°C are also available. If p.v.c. is exposed to sunlight it may be degraded by ultra-
violet radiation. If it is in contact with absorbent materials, the plasticiser may be 'leached
out' making the p.v.c. hard and brittle.

LSF (Low smoke and fume)

Materials which have reduced smoke and corrosive gas emissions in fire compared with
p.v.c. have been available for some years. They are normally used as sheathing
compounds over XLPE or LSF insulation, and can give considerable safety advantages in
situations where numbers of people may have to be evacuated in the event of fire.

Thermosetting (XLPE)
Gross-linked polyethylene (XLPE) is a thermosetting compound which has better
electrical properties than p.v.c. and is therefore used for medium- and high-voltage
applications. It has more resistance to deformation at higher temperatures than p.v.c.,
which it is gradually replacing. It is also replacing PILC in some applications.
Thermosetting insulation may be used safely with conductor temperatures up to 90°C
thus increasing the useful current rating, especially when ambient temperature is high. A
LSF (low smoke and fume) type of thermosetting cable is available.

Mineral
Provided that it is kept dry, a mineral insulation such as magnesium oxide is an excellent
insulator. Since it is hygroscopic (it absorbs moisture from the air) this insulation is kept
sealed within a copper sheath. The resulting cable is totally fireproof and will operate at
temperatures of up to 250°C. It is also entirely inorganic and thus non-ageing. These
cables have small diameters compared with alternatives, great mechanical strength, are
waterproof, resistant to radiation and electromagnetic pulses, are pliable and corrosion
resistant. In cases where the copper sheath may corrode, the cable is used with an overall
LSF covering, which reduces the temperature at which the cable may be allowed to
operate. Since it is necessary to prevent the ingress of moisture, special seals are used to
terminate cables. Special mineral-insulated cables with twisted cores to reduce the effect
of electromagnetic interference are available.

7.15.1 - Rotating machines

The vast majority of motors used in industry are of the three-phase squirrel-cage
induction type. Smaller motors are usually single-phase induction machines. Induction
motors have important advantages, such as robustness, minimal maintenance needs, and
self starting characteristics, but all draw very high starting currents from their supplies
{Fig 7.23}. This starting current is a short-lived transient, and may usually be ignored
when calculating cable sizes.

Fig 7.23 - Starting current of an induction motor

Although the starting current may be several times the running current, the value
depending on the machine characteristics and the connected mechanical load, its short
duration will not lead to overheating in usual circumstances. If frequent starting is a
requirement, larger supply cables may be necessary to avoid damage to insulation. A
problem could arise when fast-acting fuses or circuit breakers are used for short-circuit
protection; the high starting current may result in operation of the protective device. A
common, but unsatisfactory, remedy for this difficulty is to increase the rating of the
protective device, leading to a loss of proper overload protection. A possible solution is to
use dual rated fuses (gM types). For example, a 25M40 fuse has a continuous rating of 25
A and the operating characteristics of a 40 A fuse.

A word is necessary concerning motor ratings. Many years ago it was decided to replace
the horsepower as the unit of output power with the kilowatt. Unfortunately, the old
horsepower is a very long time in dying. Many machines still have rating plates giving
output power in horsepower. The conversion is straightforward. Since one horsepower is
the same as 746 W, horsepower is converted to kilowatts by multiplying by 0.746.

It is sometimes practice to stop a motor very quickly by feeding it with a reverse current.
When this method is provided it is important that the machine does not begin to move in
the reverse direction if this would cause danger.

Where other types of motor, such as wound rotor and commutator induction or thyristor
fed dc types are used, the cables must be suitable for carrying running currents on full
load, which will usually mean that they are large enough to carry the short duration
starting currents.

Every motor rated at 0.37 kW (0.5 horsepower) or more must be fed from a starter which
includes overload protection. Such devices have time-delay features so that they will not
trip as a result of high starting current, but will do so in the event of a small but prolonged
overload. They have the advantages over fuses and single-pole circuit breakers that all
three lines of a three-phase system are tripped by an overload in any one of them. If only
one line were broken, the resulting 'single-phasing' operation of the motor could cause it
to overheat.

It is often necessary to provide a means to prevent automatic restarting after failure of the
supply. For example, if the supply to a machine shop fails, the machine operators are
likely to use the enforced break in production to clean and service their machines. If so,
when the supply is restored, the presence of hands, brushes, tools, etc. in the machines
when they automatically restart would cause serious danger. The necessary 'no-volt
protection' is obtained by using a starter of the type whose circuit is shown in {Fig 7.24}.
The coil is fed through the 'hold-in' contacts, which open when the supply fails; the motor
can then be operated only by pressing the 'start' button. This requirement does not apply
to protected motors which are required to restart automatically after mains failure.
Examples are motors supplying refrigeration and pumping plants. It is important that
lock-off stop buttons are not used as a means of isolation.

Fig 7.24 - Direct-on-line starter

 5.9.2 - The principle of the residual current device

The RCD is a circuit breaker which continuously compares the current in the phase with
that in the neutral. The difference between the two (the residual current) will he flowing
to earth, because it has left the supply through the phase and has not returned in the
neutral (see {Fig 5.22}). There will always be some residual current in the insulation
resistance and capacitance to earth, but in a healthy circuit such current will he low,
seldom exceeding 2 mA.

Fig 5.22 The meaning of the term residual current

The purpose of the residual current device is to monitor the residual current and to switch
off the circuit quickly if it rises to a preset level. The arrangement of an RCD is shown in
simplified form in {Fig 5.23}. The main contacts are closed against the pressure of a
spring, which provides the energy to open them when the device trips. Phase and neutral
currents pass through identical coils wound in opposing directions on a magnetic circuit,
so that each coil will provide equal but opposing numbers of ampere turns when there is
no residual current. The opposing ampere turns will cancel, and no magnetic flux will be
set up in the magnetic circuit.

Residual earth current passes to the circuit through the phase coil but returns through the
earth path, thus avoiding the neutral coil, which will therefore carry less current. This
means that phase ampere turns exceed neutral ampere turns and an alternating magnetic
flux results in the core. This flux links with the search coil, which is also wound on the
magnetic circuit, inducing an e.m.f. into it. The value of this e.m.f. depends on the
residual current, so it will drive a current to the tripping system which depends on the
difference between phase and neutral currents. When the amount of residual current, and
hence of tripping current, reaches a pre-determined level, the circuit breaker trips,
opening the main contacts and interrupting the circuit.

For circuit breakers operating at low residual current values, an amplifier may be used in
the trip circuit. Since the sum of the currents in the phases and neutral of a three-phase
supply is always balanced, the system can be used just as effectively with three-phase
supplies. In high current circuits, it is more usual for the

Fig 5.23 Residual current circuit breaker

phase and neutral conductors to simply pass through the magnetic core instead of round
coils wound on it.

Operation depends on a mechanical system, which could possibly become stiff when old
or dirty. Thus, regular testing is needed, and the RCD is provided with a test button
which provides the rated level of residual current to ensure that the circuit breaker will
operate. All RCDs are required to display a notice which draws attention to the need for
frequent testing which can be carried out by the user, who presses a test button, usually
marked T. {Table 5.10} shows the required notice.

This installation, or part of it, is protected by a device which

automatically switches off the supply if an earth fault develops.
Test quarterly by pressing the button marked 'T' or 'Test'. The device
should switch off the supply, and should then be switched on to restore
the supply. If the device does not switch off the supply when the button
is pressed, seek expert advice

Table 5.10 - Periodic test notice for residual current device

The test circuit is shown in {Fig 5.23}, and provides extra current in the phase coil when
the test button is pressed. This extra current is determined by the value of the resistor R.

There are currently four basic types of RCD. Class AC devices are used where the
residual current is sinusoidal - this is the normal type which is in the most wide use. Class
A types are used where the residual current is sinusoidal and/or includes pulsating direct
currents - this type is applied in special situations where electronic equipment is used.
Class B is for specialist operation on pure direct current or on impulse direct or
alternating current. Class S RCDs have a built-in time delay to provide discrimination
(see below).

It must be understood that the residual current is the difference between phase and neutral
currents, and that the current breaking ability of the main contacts is not related to the
residual operating current value, There is a widely held misunderstanding of this point,
many people thinking that the residual current setting is the current breaking capability of
the device. It is very likely that a device with a breaking capacity of 100 A may have a
residual operating current of only 30 mA.

There are cases where more than one residual current device is used in an installation; for
example, a complete installation may be protected by an RCD rated at 100 mA whilst a
socket intended for equipment outdoors may be protected by a 30 mA device.
Discrimination of the two devices then becomes important. For example, if an earth fault
giving an earth current of 250 mA develops on the equipment fed by the outdoor socket,
both RCDs will carry this fault current, and both will become unbalanced. Since the fault
is higher than the operating current of both devices, both will have their trip systems
activated. It does not follow that the device with the smaller operating current will open
first, so it is quite likely that the 100 mA device will operate, cutting off the supply to the
complete installation even though the fault was on a small part of it. This is a lack of
discrimination between the residual current devices. To ensure proper discrimination, the
device with the larger operating current has a deliberate delay built into its operation. It is
called a time delayed RCD.

5.9.1 - Why do we need residual current devices?

{5.3} has stressed that the standard method of protection is to make sure that an earth
fault results in a fault current high enough to operate the protective device quickly so
that fatal shock is prevented. However, there are cases where the impedance of the
earth-fault loop, or the impedance of the fault itself, are too high to enable enough
fault current to flow. In such a case, either:

1. - current will continue to flow to earth, perhaps generating enough heat to start a
fire, or

2. - metalwork which is open to touch may be at a high potential relative to earth,

resulting in severe shock danger.

Either or both of these possibilities can be removed by the installation of a residual

current device (RCD).

In recent years there has been an enormous increase in the use of initials for residual
current devices of all kinds. The following list, which is not exhaustive, may be
helpful to readers:

RCD residual current device

 RCCD residual current operated circuit breaker
SRCD socket outlet incorporating an RCD
PRCD portable RCD, usually an RCD incorporated into a plug
RCBO an RCCD which includes overcurrent protection
SRCBO a socket outlet incorporating an RCBO

5.8.4 - Electrical separation

Safety from shock can sometimes be ensured by separating a system completely from
others so that there is no complete circuit through which shock current could flow. It
follows that the circuit must be small to ensure that earth impedance's are very high and
do not offer a path for shock current (see {Fig 5.3(b)}). The source of supply for such a
circuit could be a battery or a generating set, but is far more likely to be an isolating
transformer with a secondary winding providing no more than 500 V. Such a transformer
must comply with BS EN 60742, having a screen between its windings and a secondary
winding which has no connection to earth.

There must be no connection to earth and precautions must he taken to ensure, as far as
possible, that earth faults will not occur. Such precautions would include the use of
flexible cords without metallic sheaths, using double insulation, making sure that flexible
cords are visible throughout their length of run, and so on. Perhaps the most common
example of a separated circuit is the bathroom transformer unit feeding an electric shaver.
By breaking the link to the earthed supply system using the double wound transformer,
there is no path to earth for shock current (see {Fig 5.21}).

Fig 5.21 Bathroom shaver socket to BS EN 60742

5.7.1 - What is earthed concentric wiring?

This is the TN-C system {5.2.5} where a combined neutral and earth (PEN) conductor is
used throughout the installation as well as for the supply. The PEN conductor is the
sheath of a cable and therefore is concentric with (totally surrounds) the phase
conductor(s). The system is unusual, but where employed almost invariably uses mineral
insulated cable, the metallic copper sheath being the combined neutral and earth
conductor.

5.7.2 - Requirements for earthed concentric wiring

Earthed concentric wiring may only be used under very special conditions, which usually
involve the use of a private transformer supply or a private generating plant. Since there
is no separate path for earth currents, it follows that residual current devices (RCDs) will
not be effective and therefore must not be used. The cross-sectional area of the sheath
(neutral and earth conductor) of a cable used in such a system must never be less than 4
mm² copper, or 16 mm² aluminium or less than the inner core for a single core cable. All
multicore copper mineral insulated cables comply with this requirement, even a I mm²
two core cable having the necessary sheath cross-sectional area. However, only single
core cables of 6 mm² and below may be used. The combined protective and neutral
conductors (sheaths) of such cables must not serve more than one final circuit.

Wherever a joint becomes necessary in the PEN conductor, the contact through the
normal sealing pot and gland is insufficient; an extra earth tail must be used as shown in
{Fig 5.19}. If it becomes necessary to separate the neutral and protective conductors at
any point in an installation, they must not be connected together again beyond that point.

Fig 5.19 Earth tail seal for use in earthed concentric wiring

5.8.1 - Class II equipment

Class H equipment has reinforced or double insulation. As well as the basic insulation for
live parts, there is a second layer of insulation, either to prevent contact with exposed
conductive parts or to make sure that there can never be any contact between such parts
and live parts. The outer case of the equipment need not be made of insulating material; if
protected by double insulation, a metal case will not present any danger. It must never be
connected with earth, so connecting leads are two-core, having no protective conductor.
The symbol for a double-insulated appliance is shown in {Fig 5.20}.
 Fig 5.20 British Standard symbol for double insulation

To make sure that the double insulation is not impaired, it must not be pierced by
conducting parts such as metal screws. Nor must insulating screws be used, because there
is the possibility that they will be lost and will be replaced by metal screws. Any holes in
the enclosure of a double Insulated appliance, such as those to allow ventilation, must be
so small that fingers cannot reach live parts (IP2X protection). Class II equipment must
be installed and fixed so that the double insulation will never be impaired, and so that
metalwork of the equipment does not come into contact with the protective system of the
main installation. Where the whole of an installation is comprised of Class II equipment,
so that there is no protective system installed, the situation must be under proper
supervision to make sure that no changes are made which will introduce earthed parts.
 Fig 4.3 - 150°C rubber-insulated and braided flexible cord used
for the final connection to a luminaire

4.3.12 - Harmonic currents and neutral conductors

A perfectly balanced three-phase system (one with all three phase loads identical in all
respects) has no neutral current and thus has no need of a neutral conductor. This is often
so with motors, which are fed through three core cables in most cases.

Many three-phase loads are made up of single-phase loads, each connected between one
line and neutral. It is not likely in such cases that the loads will be identical, so the neutral
will carry the out-of-balance current of the system. The greater the degree of imbalance,
the larger the neutral current.

Some three-phase four-core cables have a neutral of reduced cross-section on the

assumption that there will be some degree of balance. Such a cable must not he used
unless the installer is certain that severe out-of-balance conditions will never occur.
Similar action must be taken with a three-phase circuit wired in single-core cables. A
reduced neutral conductor may only be used where out-of-balance currents will be very
small compared to the line currents.

A problem is likely to occur in systems which generate significant third harmonic

currents. Devices such as discharge lamp ballasts and transformers on low load distort the
current waveform. Thus, currents at three times normal frequency (third harmonics) are
produced, which do not cancel at the star point of a three-phase system as do normal
frequency currents, but add up, so that the neutral carries very heavy third harmonic
currents. For this reason, it is important not to reduce the cross-sectional area of a neutral
used to feed discharge lamps (including fluorescent lamps).

In some cases the neutral current may be considerably larger than the phase currents.
Where the load concerned is fed through a multi-core cable, it may be prudent to use
five-core (or even six-core) cables, so that two (or three) conductors may be used in
parallel for the neutral.

In some cases it may be necessary to insert overload protection in a neutral conductor.

Such protection must be arranged to open all phase conductors on operation, but not the
neutral. This clearly indicates the use of a special circuit breaker.

It is very important that the neutral of each circuit is kept quite separate from those of
other circuits. Good practice suggests that the separate circuit neutrals should be
connected in the same order at the neutral block as the corresponding phase conductors at
the fuses or circuit breakers.

5.1.1 - What is earthing?

The whole of the world may be considered as a vast conductor which is at reference
(zero) potential. In the UK we refer to this as 'earth' whilst in the USA it is called
'ground'. People are usually more or less in contact with earth, so if other parts which are
open to touch become charged at a different voltage from earth a shock hazard exists (see
{3.4}). The process of earthing is to connect all these parts which could become charged to
the general mass of earth, to provide a path for fault currents and to hold the parts as
close as possible to earth potential. In simple theory this will prevent a potential
difference between earth and earthed parts, as well as permitting the flow of fault current
which will cause the operation of the protective systems.

The standard method of tying the electrical supply system to earth is to make a direct
connection between the two. This is usually carried out at the supply transformer, where
the neutral conductor (often the star point of a three-phase supply) is connected to earth
using an earth electrode or the metal sheath and armouring of a buried cable. {Figure 5.1}
shows such a connection. Lightning conductor systems must be bonded to the installation
earth with a conductor no larger in cross-sectional area than that of the earthing
conductor.

Fig 5.1 - Three-phase delta/star transformer

showing earthing arrangements

5.1.2 - The advantages of earthing

The practice of earthing is widespread, but not all countries in the world use it.
There is certainly a high cost involved, so there must be some advantages. In fact
there are two. They are:

1. - The whole electrical system is tied to the potential of the general mass of earth and
cannot 'float' at another potential. For example, we can be fairly certain that the neutral of
our supply is at, or near, zero volts (earth potential) and that the phase conductors of our
standard supply differ from earth by 240 volts.

2. - By connecting earth to metalwork not intended to carry current (an extraneous

conductive part or a an exposed conductive part) by using a protective conductor, a path
is provided for fault current which can be detected and, if necessary, broken. The path for
this fault current is shown in {Fig 5.2}.

Fig 5.2 Path for earth fault current (shown by arrows)

5.1.3 - The disadvantages of earthing

The two important disadvantages are:

1. - Cost: the provision of a complete system of protective conductors, earth electrodes,

etc. is very expensive.

2. - Possible safety hazard: It has been argued that complete isolation from earth will
prevent shock due to indirect contact because there is no path for the shock current to
return to the circuit if the supply earth connection is not made (see {Fig 5.3(a)}). This
approach, however, ignores the presence of earth leakage resistance (due to imperfect
insulation) and phase-to-earth capacitance (the insulation behaves as a
dielectric). In many situations the combined impedance due to insulation resistance and
earth capacitive reactance is low enough to allow a significant shock current
(see {Fig 5.3(b)}).

Fig 5.3 - Danger in an unearthed system

a) apparent safety: no obvious path for shock current

b) actual danger: shock current via stray resistance and capacitance

5.2.1 -System classification

The electrical installation does not exist on its own; the supply is part of the overall
system. Although Electricity Supply Companies will often provide an earth terminal, they
are under no legal obligation to do so. As far as earthing types are concerned, letter
classifications are used.

The first letter indicates the type of supply earthing.

T -indicates that one or more points of the Supply are directly earthed (for
example, the earthed neutral at the transformer).
I -indicates either that the supply system is not earthed at all, or that the earthing
includes a deliberately-inserted impedance, the purpose of which is to limit fault current.
This method is not used for public supplies in the UK.

The second letter indicates the earthing arrangement in the installation.

T - all exposed conductive metalwork is connected directly to earth.
N - all exposed conductive metalwork is connected directly to an earthed supply
conductor provided by the Electricity Supply Company.

The third and fourth letters indicate the arrangement of the earthed supply conductor
system.
S - neutral and earth conductor systems are quite separate.
C - neutral and earth are combined into a single conductor.
A number of possible combinations of earthing systems in common use is indicated in the
following subsections.

Protective conductor systems against lightning need to be connected to the installation

earthing system to prevent dangerous potential differences. Where a functional earthing
system is in use, the protective requirements of the earthing will take precedence over the
functional requirements.

5.2.2 - TT systems

This arrangement covers installations not provided with an earth terminal by the
Electricity Supply Company. Thus it is the method employed by most (usually rural)
installations fed by an overhead supply. Neutral and earth (protective) conductors must
be kept quite separate throughout the installation, with the final earth terminal connected
to an earth electrode (see {5.5}) by means of an earthing conductor.

Effective earth connection is sometimes difficult. Because of this, socket outlet circuits
must be protected by a residual current device (RCD) with an operating current of 30 mA
{5.9}. {Fig 5.4} shows the arrangement of a TT earthing system.

Fig 5.4 - TT earthing system

5.2.3 - TN-S system

This is probably the most usual earthing system in the UK, with the Electricity Supply
Company providing an earth terminal at the incoming mains position. This earth terminal
is connected by the supply protective conductor (PE) back to the star point (neutral) of
the secondary winding of the supply transformer, which is also connected at that point to
an earth electrode. The earth conductor usually takes the form of the armour and sheath
(if applicable) of the underground supply cable. The system is shown diagrammatically in
{Fig 5.5}.
 Fig 5.5 - TN-S earthing system

5.2.4 - TN-C-S system

In this system, the installation is TN-S, with separate neutral and protective conductors.
The supply, however, uses a common conductor for both the neutral and the earth. This
combined earth and neutral system is sometimes called the 'protective and neutral
conductor' (PEN) or the 'combined neutral and earth' conductor (CNE). The system,
which is shown diagrammatically in {Fig 5.6}, is most usually called the protective
multiple earth (PME) system, which will be considered in greater detail in {5.6}.

Fig 5.6 - TN-C-S earthing system - protective multiple earthing

5.2.5 -TN-C system

This installation is unusual, because combined neutral and earth wiring is used in both the
supply and within the installation itself. Where used, the installation will usually be the
earthed concentric system, which can only be installed under the special conditions listed
in {5.7}.

5.2.6 - IT system

The installation arrangements in the IT system are the same for those of the TT system
{5.2.2}. However, the supply earthing is totally different. The IT system can have an
unearthed supply, or one which is not solidly earthed but is connected to earth through a
current limiting impedance.

The total lack of earth in some cases, or the introduction of current limiting into the earth
path, means that the usual methods of protection will not be effective. For this reason, IT
systems are not allowed in the public supply system in the UK. An exception is in
medical situations such as hospitals. Here it is recommended that an IT system is used for
circuits supplying medical equipment that is intended to be used for life-support of
patients. The method is also sometimes used where a supply for special purposes is taken
from a private generator.

5.3.1 - Principle

The path followed by fault current as the result of a low impedance occurring between
the phase conductor and earthed metal is called the earth fault loop. Current is driven
through the loop impedance by the supply voltage.

The extent of the earth fault loop for a TT system is shown in {Fig 5.7}, and is made up
of the following labelled parts.

Fig 5.7 The earth fault loop

l. - the phase conductor from the transformer to the installation

2 . -the protective device(s) in the installation

3 . -the installation phase conductors from the intake position
to the fault

4. - the fault itself (usually assumed to have zero impedance)

5. - the protective conductor system

6. - the main earthing terminal

7. - the earthing conductor

8. - the installation earth electrode

9. - the general mass of earth

10. - the Supply Company's earth electrode

11. - the Supply Company's earthing conductor

12 . -the secondary winding of the supply transformer

For a TN-S system (where the Electricity Supply Company provides an earth terminal),
items 8 to 10 are replaced by the PE conductor, which usually takes the form of the
armouring (and sheath if there is one) of the underground supply cable.

For a TN-C-S system (protective multiple earthing) items 8 to 11 are replaced by the
combined neutral and earth conductor.

For a TN-C system (earthed concentric wiring), items 5 to 11 are replaced by the
combined neutral and earth wiring of both the installation and of the supply.

It is readily apparent that the impedance of the loop will probably be a good deal higher
for the TT system, where the loop includes the resistance of two earth electrodes as well
as an earth path, than for the other methods where the complete loop consists of metallic
conductors.

5.3.2 - The importance of loop impedance

The earth fault loop impedance can be used with the supply voltage to calculate the earth-
fault current.

IF =Uo
Zs
whereIF = fault current, A
Uo = phase voltage, V
Zs = loop impedance
For example, if a 240 V circuit is protected by a 15 A semi-enclosed fuse and has an
earth-fault loop impedance of 1.6 Ohms, the earth-fault current in the event of a zero
impedance earth fault will he:

Uo 240
IF = 150 A
= A
Zs 1.6

This level of earth-fault current will cause the fuse to operate quickly. From {Fig 3.13} the
time taken for the fuse to operate will be about 0.15 s. Any load current in the circuit will
be additional to the fault current and will cause the fuse to operate slightly more quickly.
However, such load current must not be taken into account when deciding disconnection
time, because it is possible that the load may not be connected when the fault occurs.

Note that there is no such thing as a three-phase line/earth fault, although it is possible for
three faults to occur on the three lines to earth simultaneously. As far as calculations for
fault current are concerned, the voltage to earth for standard UK supplies is always 240
V, for both single-phase and three-phase systems. Thus the Tables of maximum earth-
fault loop impedance which will be given in {5.3.4} apply both to single- and to three-
phase systems.

5.3.3 - The resistance/impedance relationship

Resistance, measured in ohms, is the property of a conductor to limit the flow of current
through it when a voltage is applied.

I= U
R
where I = current, A
U
applied voltage. V
=
R circuit resistance,
= Ohms

Thus, a voltage of one volt applied to a one ohm resistance results in a current of one
ampere.

When the supply voltage is alternating, a second effect, known as reactance (symbol X) is
to be considered. It applies only when the circuit includes inductance and/or capacitance,
and its value, measured in ohms, depends on the frequency of the supply as well as on the
values of the inductance and/or the capacitance concerned. For almost all installation
work the frequency is constant at 50 Hz. Thus, inductive reactance is directly
proportional to inductance and capacitive reactance is inversely proportional to
capacitance.

Xl = 2(pi)fL and Xc = 1
2(pi)fC
Xl
where inductive reactance (Ohms)
=
 Xc =capacitive reactance (Ohms)
(pi) =the mathematical constant (3.142)
f =the supply frequency (Hz)
L =circuit inductance (H)
C =circuit capacitance (F)

Resistance (R) and reactance (Xl or Xc) in series add together to produce the circuit
impedance (symbol z), but not in a simple arithmetic manner. Impedance is the effect
which limits alternating current in a circuit containing reactance as well as resistance.

Z =U
I
where Z =impedance (Ohms)
U =applied voltage (V)
I =current (A)

It follows that a one volt supply connected across a one ohm impedance results in a
current of one ampere.

When resistance and reactance are added this is done as if they were at right angles,
because the current in a purely reactive circuit is 90° out of phase with that in a purely
resistive circuit. The relationships between resistance, reactance and impedance are:

a) resistive and capacitive circuit - b) resistive and inductive circuit

Fig 5.8 Impedance diagrams

These relationships can be shown in the form of a diagram applying Pythagoras' theorem
as shown in {Fig 5.8}. The two diagrams are needed because current lags voltage in the
inductive circuit, but leads it in the capacitive. The angle between the resistance R and
the impedance Z is called the circuit phase angle, given the symbol a (Greek 'phi'). If
voltage and current are both sinusoidal, the cosine of this angle, cos a, is the circuit power
factor, which is said to be lagging for the inductive circuit, and leading for the capacitive.
In practice, all circuits have some inductance and some capacitance associated with them.
However, the inductance of cables only becomes significant when they have a cross-
sectional area of 25 mm² and greater. Remember that the higher the earth fault loop
impedance the smaller the fault current will be. Thus, if simple arithmetic is used to add
resistance and reactance, and the resulting impedance is low enough to open the
protective device quickly enough, the circuit will be safe. This is because the
Pythagorean addition will always give lower values of impedance than simple addition.

For example, if resistance is 2 Ohms and reactance 1 Ohm, simple arithmetic addition
gives

Z = R + X – 2 + 1 = 3 Ohms
and correct addition gives
Z = (R² + X²)
= (2² + 1²) =  5 = 2.24 Ohms

If 3 Ohms is acceptable, 2.24 Ohms will allow a larger fault current to flow which will
operate the protective device more quickly and is thus even more acceptable.

5.3.4 - Earth-fault loop impedance values

The over-riding requirement is that sufficient fault current must flow in the event of an
earth fault to ensure that the protective device cuts off the supply before dangerous shock
can occur. For normal 240 V systems, there are two levels of maximum disconnection
time. These are:

For socket outlet circuits where equipment could be tightly grasped: 0.4 s

For fixed equipment where contact is unlikely to be so good: 5 s

The maximum disconnection time of 5 s also applies to feeders and sub-mains.

It must be appreciated that the longest disconnection times for protective devices, leading
to the longest shock times and the greatest danger, will be associated with the lowest
levels of fault current, and not, as is commonly believed, the highest levels.

Where the voltage is other than 240 V, [Table 41A] gives a range of disconnection times
for socket outlet circuits, of which the lowest is 0.1 s for voltages exceeding 400 V.

In general, the requirement is that if a fault of negligible impedance occurs between a

phase and earth, the earth-fault loop impedance must not be greater than the value
calculated from..

Zs < Uo
Ia

where Zs = the earth fault loop impedance (Ohms)

 Uo = the system voltage to earth(V)
Ia = the current causing automatic disconnection
(operation of the protective device) in the required time [A]).

The earth fault loop values shown in [Tables 5.1, 5.2 and 5.4] depend on the supply voltage
and assume, as shown in the Tables, a value of 240 V. Whilst it would appear that 240 V
is likely to be the value of the supply voltage in Great Britain for the foreseeable future, it
is not impossible that different values may apply. In such a case, the tabulated value for
earth fault loop impedance should be modified using the formula:-

Zs =Zt x U
U240

where Zs =is the earth fault loop impedance required for safety
Zt =is the tabulated value of earth fault loop impedance
U =is the actual supply voltage
U240 =is the supply voltage assumed in the Table.

As an alternative to this calculation, a whole series of maximum values of earth fault loop
impedance is given in {Table 5.1} (from [Table 41B]) for disconnection within 0.4 s. The
reader should not think that these values are produced in some mysterious way - all are
easily verified using the characteristic curves {Figs 3.13 to 3.19}.

For example, consider a 20 A HRC fuse to BS88 used in a 240 V system. The fuse
characteristic is shown in {Fig 3.15}, and indicates that disconnection in 0.4 s requires a
current of about 130 A. It is difficult (if not impossible) to be precise about this value of
current, because it is between the 100 A and 150 A current graduations.

Using these values,

Zs
Uo = 240 Ohms = 1.84 Ohms
=
Ia 130

Reference to {Table 5.1} shows that the stated value is 1.8 Oh,s, the discrepancy being due
to the difficulty in reading the current with accuracy. {Tables 5.1 and 5.2} give maximum
earth-fault loop impedance values for fuses and for miniature circuit breakers to give a
minimum disconnection time of 0.4 s in the event of a zero impedance fault from phase
to earth.

The reason for the inclusion of fixed equipment as well as distribution circuits in {Table
5.2} will become apparent later in this sub-section.

Table 5.1 - Maximum earth-fault loop impedance for 240 V socket

---------------- outlet circuits protected by fuses
Fuse rating (A) Maximum earth-fault loop impedance (Ohms)
Cartridge Semi-enclosed
- Cartridge BS 1361
BS 88 BS3036
 5 - 10.9 10.0
6 8.89 - -
10 5.33 - -
15 - 3.43 2.67
20 1.85 1.78 1.85
30 - 1.20 1.14
32 1.09 - -
40 0.86 - -
45 - 0.60 0.62

Table 5.2 - Maximum earth-fault loop impedance for 240 V circuits

-----------------protected by miniature circuit breakers to give compliance
---------------- with 0.4 s disconnection time
- Maximum earth-fault loop impedance (Ohms)
MCB
Device MCB MCB MCB MCB
type 3
rating (A) type 1 type 2 type B type D
and type C
5 12.00 6.86 4.80 - 2.40
6 10.00 5.71 4.00 8.00 2.00
10 6.00 3.43 2.40 4.80 1.20
15 4.00 2.29 1.60 - 0.80
16 3.75 2.14 1.50 3.00 0.75
20 3.00 1.71 1.20 2.40 0.60
25 2.40 1.37 0.96 1.92 0.48
30 2.00 1.14 0.80 - 0.40
32 1.88 1.07 0.75 1.50 0.38
40 1.5 0.86 0.60 1.20 0.30

The severity of the electric shock received when there is a phase to earth fault (indirect
contact) depends entirely on the impedance of the circuit protective conductor. We saw in
{3.4.3} and {Fig 3.8} how the volt drop across the protective conductor is applied to the
person receiving the shock. Since this volt drop is equal to fault current times protective
conductor impedance, if the protective conductor has a lower impedance the shock
voltage will he less. Thus it can be sustained for a longer period without extreme danger.

Socket outlet circuits can therefore have a disconnection time of up to 5 s provided that
the circuit protective conductor impedance's are no higher than shown in {Table 5.3} for
various types of protection.
The reasoning behind this set of requirements becomes clearer if we take an example.
{Table 5.3} shows that a 40 A cartridge fuse to BS 88 must have an associated protective
conductor impedance of no more than 0.29 Ohms if it is to comply. Now look at the
time/current characteristic for the fuse {Fig 3.15} from which we can see that the current
for operation in 5 s is about 170 A. The maximum volt drop across the conductor (the
shock voltage) is thus 170 x 0.29 or 49.3 V.

Table 5.3 - Maximum impedance of circuit protective conductors to

---------------- allow 5 s disconnection time for socket outlets
- Maximum impedance of circuit protective conductor
Fuse Fuse Fuse MCB
Rating MCB MCB MCB MCB
BS 88 BS BS type 3 &
(A) type 1 type 2 type B type D
1361 3036 C
5 - 3.25 3.25 2.50 1.43 1.00 - 0.50
6 2.48 - - 2.08 1.19 0.83 1.67 0.42
10 1.48 - - 1.25 0.71 0.50 1.00 0.25
15 - 0.96 0.96 0.83 0.48 0.33 - -
16 0.83 - - 0.78 0.45 0.31 0.63 0.16
20 0.55 0.55 0.63 0.63 0.36 0.25 0.50 0.12
25 0.43 - - - - - - 0.10
30 - 0.36 0.43 0.42 0.24 0.17 - -
32 0.34 - - 0.39 0.22 0.16 0.31 0.08
40 0.26 - - 0.31 0.18 0.13 0.25 0.06
45 - 0.18 0.24 0.28 0.16 0.11 0.22 0.06

Table 5.4 - Maximum earth-fault loop impedance for 240 V fixed

---------------- equipment distribution circuits protected by fuses
- Maximum earth-fault loop impedance
Device rating Cartridge Cartridge Semi-enclosed
(A) BS 88 BS 1361 BS 3036
5 - 17.1 -
6 14.1 - -
10 7.74 - -
15 - 5.22 5.58
16 4.36 - -
20 3.04 2.93 4.00
 30 - 1.92 2.76
32 1.92 - -
40 1.41 - -
45 - 1.00 1.66
50 1.09 - -

Application of the same reasoning to all the figures gives shock voltages of less than 50
V. This limitation on the impedance of the CPC is of particular importance in TT systems
where it is likely that the resistance of the earth electrode to the general mass of earth will
be high.

The breaking time of 5 s also applies to fixed equipment, so the earth-fault loop
impedance values can be higher for these circuits, as well as for distribution circuits. For
fuses, the maximum values of earth-fault loop impedance for fixed equipment are given
in {Table 5.4}.

No separate values are given for miniature circuit breakers. Examination of the
time/current characteristics {Figs 3.16 to 3.19} will reveal that there is no change at all in
the current causing operation between 0.4 s and 5 s in all cases except the Type 1. Here,
the vertical characteristic breaks off at 4 s, but this makes little difference to the
protection. In this case, the values given in {Table 5.2} can be used for fixed equipment as
well as for socket outlet circuits. An alternative is to calculate the loop impedance as
described above.

5.3.5 - Protective conductor impedance

It has been shown in the previous sub-section how a low-impedance protective conductor
will provide safety from shock in the event of a fault to earth. This method can only be
used where it is certain that the shock victim can never be in contact with conducting
material at a different potential from that of the earthed system in the zone he occupies.
Thus, all associated exposed or extraneous parts must be within the equipotential zone
(see {5.4}). When overcurrent protective devices are used as protection from electric
shock, the protective conductor must be in the same wiring system as, or in close
proximity to, the live conductors. This is intended to ensure that the protective conductor
is unlikely to he damaged in an accident without the live conductors also being cut.

{Figure 5.9} shows a method of measuring the resistance of the protective conductor,
using a line conductor as a return and taking into account the different cross-sectional
areas of the phase and the protective conductors.
 Fig 5.9 - Measurement of protective conductor resistance

Taking the cross-sectional area of the protective conductor as Ap and that of the line
(phase or neutral) conductor as Al , then

Rp = resistance reading x Al
Al + Ap

For example, consider a reading of 0.72 Ohms obtained when measuring a circuit in the
way described and having 2.5 mm² line conductors and a 1.5 mm² protective conductor.
The resistance of the protective conductor is calculated from:

Rp = R x Al = 0.72 x 2.5 Ohms

Al + Ap 2.5 + 1.5

= 0.72 x 2.5 Ohms = 0.45 Ohms

4.0

5.4.1 - Earthing conductors

The earthing conductor is commonly called the earthing lead. It joins the installation
earthing terminal to the earth electrode or to the earth terminal provided by the Electricity
Supply Company. It is a vital link in the protective system, so care must be taken to see
that its integrity will be preserved at all times. Aluminium conductors and cables may
now be used for earthing and bonding, but great care must be taken when doing so to
ensure that there will be no problems with corrosion or with electrolytic action where
they come into contact with other metals.

Where the final connection to the earth electrode or earthing terminal is made there must
be a clear and permanent label Safety Electrical Connection - Do not remove (see {Fig
5.17}). Where a buried earthing conductor is not protected against mechanical damage but
is protected against corrosion by a sheath, its minimum size must be 16 mm² whether
made of copper or coated steel. If it has no corrosion protection, minimum sizes for
mechanically unprotected earthing conductors are 25 mm² for copper and 50 mm² for
coated steel.

If not protected against corrosion the latter sizes again apply, whether protected from
mechanical damage or not.

Earthing conductors, as well as protective and bonding conductors, must be protected

against corrosion. Probably the most common type of corrosion is electrolytic, which is
an electro-chemical effect between two different metals when a current passes between
them whilst they are in contact with each other and with a weak acid. The acid is likely to
be any moisture which has become contaminated with chemicals carried in the air or in
the ground. The effect is small on ac supplies because any metal removed whilst current
flows in one direction is replaced as it reverses in the next half cycle. For dc Systems,
however, it will be necessary to ensure that the system remains perfectly dry (a very
difficult task) or to use the 'sacrificial anode' principle.

A main earth terminal or bar must be provided for each installation to collect and connect
together all protective and bonding conductors. It must be possible to disconnect the
earthing conductor from this terminal for test purposes, but only by the use of a tool. This
requirement is intended to prevent unauthorised or unknowing removal of protection.

5.4.4 - Protective conductor cross-section assessment

A fault current will flow when an earth fault occurs. If this current is large enough to
operate the protective device quickly, there is little danger of the protective conductor and
the exposed conductive parts it connects to earth being at a high potential to earth for
long enough for a dangerous shock to occur. The factors determining the fault current are
the supply voltage and the earth-fault loop impedance (see{5.3}).

The earth fault results in the protective conductors being connected in series across the
supply voltage {Fig 5.16}. The voltage above earth of the earthed metalwork (the voltage
of the junction between the protective and phase conductors) at this time may become
dangerously high, even in an installation complying with the Regulations. The people
using the installation will be protected by the ability of the fuse or circuit breaker in a
properly designed installation to cut off the supply before dangerous shock damage
results.
 Fig 5.16 -The effect of protective conductor resistance on shock voltage

a) effective resistance of a ring circuit protective conductor

b) potential differences across healthy protective conductor in the event of an earth fault

Table 5.7 - Main earthing and main equipotential bonding conductor

----------------- sizes for TN-S and TN-C-S supplies
Phase Earthing conductor Main
Main equipotential
conductor (or (not buried or equipotential
bonding conductor
neutral for PME protected against bonding
for PME supplies
supplies ) mechanical damage) conductor
csa mm² csa mm² csa mm² csa mm²
4 4 6 10
6 6 6 10
10 10 6 10
16 16 10 10
25 16 10 10
35 16 10 10
50 25 16 16
70 35 16 25

Remember that lower fault levels result in a longer time before operation of the protective
device. Since the cross-sectional area of the protective conductor will usually be less than
that of live conductors, its temperature, and hence its resistance, will become higher
during the fault, so that the shock voltage will be a higher proportion of the supply
potential (see {Fig 5.16}).

{Fig 5.16} shows the circuit arrangements, with some typical phase- and
protective-conductor resistances. In this case, a shock voltage of 140 V will be applied to
a person in contact with earthed metal and with the general mass of earth. Thus, the
supply must he removed very quickly. The actual voltage of the shock depends directly
on the relationship between the phase conductor resistance and the protective conductor
resistance. If the two are equal, exactly half the supply voltage will appear as the shock
voltage.

For socket outlet circuits, where the shock danger is highest, the maximum protective
conductor resistance values of {Table 5.3} will ensure that the shock voltage never exceeds
the safe value of 50 V. If the circuit concerned is in the form of a ring, one quarter of the
resistance of the complete protective conductor round the ring must not be greater than
the {Table 5.3} figure. The reason for this is shown in {Fig 5.16(a)}. This assumes that the
fault will occur exactly at the mid point of the ring. If it happens at any other point,
effective protective conductor resistance is lower, and safer, than one quarter of the total
ring resistance.
{Table 5.7} allows selection (rather than calculation) of sizes for earthing and bonding
conductors. The rules applying to selection are:

For phase conductors up to 16 mm², the protective conductor has the same size as the
phase conductor.

For phase conductors from 16 mm² to 35 mm², the protective conductor must be 16 mm²

For phase conductors over 35 mm², the protective conductor must have at least half the
c.s.a. of the phase conductor. The minimum cross-sectional area of a separate CPC is 2.5
mm² if mechanically protected and 4mm² if not.

Note that Regional Electricity Companies may require a minimum size of earthing
conductor of 16 mm² at the origin of the installation. Always consult them before
designing an installation.

5.4.5 - Protective conductor cross-section calculation

The c.s.a. of the circuit protective conductor (c.p.c.) is of great importance since

the level of possible shock in the event of a fault depends on it (as seen in {5.4.4}).

Safety could always be assured if we assessed the size using {Table 5.7} as a basis.

However, this would result in a more expensive installation than necessary because we
would often use protective conductors which are larger than those found to be acceptable
by calculation. For example. twin with cpc insulated and sheathed cables larger than 1
mm² would be ruled out because in all other sizes the CPC is smaller than required by
{Table 5.7}.

In very many cases, calculation of the CPC size will show that a smaller size than that
detailed in {5.4.4} is perfectly adequate. The formula to be used is:

S= (Ia²t)
k
S is the minimum protective conductor cross-
where
sectional area (mm2)
Ia is the fault current (A)
t is the opening time of the protective device (s)
k is a factor depending on the conductor material
and insulation, and the initial and maximum insulation
temperatures.

This is the same formula as in {3.7.3}, the adiabatic equation, but with a change in the
subject. To use it, we need to have three pieces of information, Ia, t and k.

1) To find Ia
 Since Ia = Uo we need values for Ia = uo and Zs
Zs

is simply the supply voltage, which in most

Uo
cases will be 240V.
is the earth-fault loop impedance
Zs assuming that the fault has zero
impedance.

Since we must assume that we are at the design stage, we cannot measure the loop
impedance and must calculate it by adding the loop impedance external to the installation
(Ze) to the resistance of the conductors to the furthest point in the circuit concerned. This
technique was used in {5.3.6}.

Thus, Zs = Ze + R1 + R2 where R1 and R2 are the resistances of the phase and protective
conductors respectively from {Table 5.5}.

2) To find t
We can find t from the time/current characteristics of {Figs 3.13 to 3.19} using the value of
Ia already calculated above. For example, if the protective device is a 20 A miniature
circuit breaker type I and the fault current is 1000 A, we shall need to consult {Fig 3.16},
when we can read off that operation will be in 0.01 s (10 ms). (It is of interest here to
notice that if the fault current had been 80 A the opening time could have been anything
from 0.04 s to 20 s,so the circuit would not have complied with the required opening
times).

3) To find k
k is a constant, which we cannot calculate but must obtain from a suitable table of values.
Some values of k for typical protective conductors are given in {Table 5.8}.

It is worth pointing out here that correctly installed steel conduit and trunking will always
meet the requirements of the Regulations in terms of protective conductor impedance.

Although appearing a little complicated, calculation of acceptable protective conductor

size is worth the trouble because it often allows smaller sizes than those shown in {Table
5.7}.

Table 5.8 - Values of k for protective conductors

Initial
Final temp Conductor
Nature of protective conductor temp. K
(°C) material
(°C)
p.v.c. insulated, not in cable
30 160 Copper 143
or bunched
- 30 160 Aluminium 95
- 30 160 Steel 52
 -
p.v.c. insulated, in cable
70 160 Copper 115
or bunched
- 70 160 Aluminium 76
-
Steel conduit or trunking 50 160 Steel 47
-
Bare conductor 30 200 Copper 159
- 30 200 Aluminium 105
- 30 200 Steel 58

Example 5.2
A load takes 30 A from a 240 V single phase supply and is protected by a 32 A HBC fuse
to BS 88. The wiring consists of 4 mm² single core p.v.c. insulated cables run in trunking,
the length of run being 18 m. The earth-fault loop impedance external to the installation
is assessed as 0.7 Ohms. Calculate the cross-sectional area of a suitable p.v.c. sheathed
protective conductor.

This is one of those cases where we need to make an assumption of the answer to the
problem before we can solve it. Assume that a 2.5 mm² protective conductor will be
acceptable and calculate the combined resistance of the phase and protective conductors
from the origin of the installation to the end of the circuit. From {Table 5.5}, 2.5 mm²
cable has a resistance of 7.4 mohms/m and 4 mm² a resistance of 4.6 mOhms/m. Both
values must be multiplied by 1.2 to allow for increased resistance as temperature rises
due to fault current.

Thus, R1 + R2 = (7.4 + 4.6) x 1.2 x 18 Ohms = 12.0 x 1.2 x 18 = 0.26 Ohms

1000 1000

This conductor resistance must be added to external loop impedance to give the total
earth-fault loop impedance.

Zs = Ze + Rl + R2 = 0.7+ 0.26 Ohms = 0.96 Ohms

We can now calculate the fault current:

la = Uo = 240 = 250A
Zs 0.96

Next we need to find the operating time for a 32 A BS 88 fuse carrying 250 A.
Examination of {Fig 3.15} shows that operation will take place after 0.2 s.
Finally, we need a value for k. From {Table 5.8} we can read this off as 115, because the
protective conductor will be bunched with others in the trunking.

We now have values for Ia, t and k so we can calculate conductor size.

S= (Ia²t) = (250² x 0.02) mm² = 0.97 mm²

k 115

This result suggests that a 1.0 mm² protective conductor will suffice. However, it may he
dangerous to make this assumption because the whole calculation has been based on the
resistance of a 2.5 mm² conductor. Let us start again assuming a 1.5 mm² protective
conductor and work the whole thing through again.

The new size protective conductor has a resistance of 18.1 mOhms/m, see {Table 5.5}, and
with the 4 mm² phase conductor gives a total conductor resistance, allowing for increased
temperature, of 0.491 Ohms. When added to external loop impedance this gives a total
earth-fault loop impedance of 1.191 Ohms and a fault current at 240 V of 202 A. From
{Fig 3.15} operating time will be 0.6 s. The value of k will be unchanged at 115.

S= (Ia²t) = (202² x 0.6) mm² = 1.36 mm²

k 115

Thus, a 1.5 mm² protective conductor can be used in this case. Note that if the size had
been assessed rather than calculated, the required size would be 4 mm², two sizes larger.
A point to notice here is that the disconnection time with a 1.5mm² protective conductor
is 0.6 s, which is too long for socket outlet circuits (0.4 s max.).

Example 5.3
A 240 V, 30 A ring circuit for socket outlets is 45 m long and is to be wired in 2.5 mm²
flat twin p.v.c. insulated and sheathed cable incorporating a 1.5 mm² cpc. The circuit is to
be protected by a semi-enclosed (rewirable) fuse to BS 3036, and the earth-fault loop
impedance external to the installation has been ascertained to be 0.3 Ohms. Verify that
the 1.5 mm² cpc enclosed in the sheath is adequate.

First use {Table 5.5} to find the resistance of the phase and cpc conductors. These are 7.4
mOhms/m and 12.1 mOhms/m respectively, so for a 45 m length and allowing for the
resistance increase with temperature factor of 1.2.

R1 + R2 = (7.4 + 12.1) x 1.2 x 45 Ohms = 19.5 x 1.2 x 45 Ohms = 1.05 Ohms

1000 1000

Zs = Ze + R1+ R2 = 0.3 + 1.05 Ohms = 0.3 + 0.263 Ohms = 0.563 Ohms

4 4

The division by 4 is to allow for the ring nature of the circuit.

Ia = Uo = 240 A = 426A
 Zs 0.563

We must then use the time/current characteristic of {Fig 3.13} to ascertain an operating
time of 0.10 s.

From {Table 5.8} the value of k is 115.

Then S = (Ia²t) = (426² x 0.10) mm² = 1.17 mm²

k 115

Since this value is smaller than the intended value of 1.5 mm², this latter value will be
satisfactory.

Example 5.4
A 240 V single-phase circuit is to be wired in p.v.c. insulated single core cables enclosed
in plastic conduit. The circuit length is 45 m and the live conductors are 16 mm² in cross-
sectional area. The circuit will supply fixed equipment, and is to be protected by a 63 A
HBC fuse to BS 88. The earth-fault loop impedance external to the installation has been
ascertained to be 0.58 Ohms. Calculate a suitable size for the circuit protective conductor.

With the information given this time the approach is somewhat different. We know that
the maximum disconnection time for fixed equipment is 5 s, so from the time/current
characteristic for the 63 A fuse {Fig 3.15} we can see that the fault current for
disconnection will have a minimum value of 280 A.

Thus,
Uo = 240 Ohms = 0.857 Ohms
Zs =
Ia 280

If we deduct the external loop impedance, we come to the resistance of phase and
protective conductors.

R1 +R2 = Zs-Ze = 0.857-0.58Ohms = 0.277 Ohms

Converting this resistance to the combined value of R1 and R2 per metre,

(R1 + R2) per metre = 0.277 x 1000 mOhms/m = 5.13 mOhms/m

45 x 1.2

Consulting {Table 5.5} we find that the resistance of 16 mm² copper conductor is 1.15
mOhms/m, whilst 10 mm² and 6 mm² are 1.83 and 3.08 mOhms/m respectively. Since
1.15 and 3.08 add to 4.23, which is less than 5.13, it would seem that a 6 mm² protective
conductor will be large enough. However, to be sure we must check with the adiabatic
equation.

R1 + R2 = (1.15+3.08) x 1.2 x 45 Ohms = 0.228 Ohms

1000
 Zs = Ze+(R1 +R2) = 0.58+0.228 Ohms = 0.808 Ohms

Ia = Uo = 240 A = 297 A
Zs 0.808

From {Fig 3.15} the disconnection time for a 63 A fuse carrying 297 A is found to he 3.8s.
From {Table 5.8} the value of k is 115.

Then S = (Ia²t) = (297² x 3.8) mm² = 5.03 mm²

k 115

Since 5.03 is less than 6 then a 6 mm² protective conductor will be large enough to satisfy
the requirements

5.4.6 - Unearthed metalwork

If exposed conductive parts are isolated, or shrouded in non-conducting material, or are

small so that the area of contact with a human body is limited, it is permissible not to
earth them. Examples are overhead line metalwork which is out of reach, steel
reinforcing rods within concrete lighting columns, cable clips, nameplates, fixing screws
and so on. Where areas are accessible only to skilled or instructed persons, and where
unauthorised persons are unlikely to enter due to the presence of warning notices, locks
and so on, earthing may be replaced by the provision of obstacles which make direct
contact unlikely, provided that the installation complies with the Electricity at Work
Regulations, 1989.

5.5.1 -Why must we have earth electrodes?

The principle of earthing is to consider the general mass of earth as a reference (zero)
potential. Thus, everything connected directly to it will be at this zero potential, or above
it by the amount of the volt drop in the connection system (for example, the volt drop in a
protective conductor carrying fault current). The purpose of the earth electrode is to
connect to the general mass of earth.

With the increasing use of underground supplies and of protective multiple earthing
(PME) it is becoming more common for the consumer to be provided with an earth
terminal rather than having to make contact with earth using an earth electrode.

5.5.2 - Earth electrode types

Acceptable electrodes are rods, pipes, mats, tapes, wires, plates and structural steelwork
buried or driven into the ground. The pipes of other services such as gas and water must
not be used as earth electrodes although they must be bonded to earth as described in
{5.4.3}. The sheath and armour of a buried cable may be used with the approval of its
owner and provided that arrangements can be made for the person responsible for the
installation to be told if the cable is changed, for example, for a type without a metal
sheath.

The effectiveness of an earth electrode in making good contact with the general mass of
earth depends on factors such as soil type, moisture content, and so on. A permanently-
wet situation may provide good contact with earth, but may also limit the life of the
electrode since corrosion is likely to be greater. If the ground in which the electrode is
placed freezes, there is likely to be an increase in earth resistance. In most parts of the
UK an earth electrode resistance in the range 1 Ohm to 5 Ohms is considered to be
acceptable.

The method of measuring the resistance of the earth electrode will be considered in
{8.6.1}; the resistance to earth should be no greater than 220 Ohms. The earthing
conductor and its connection to the earth electrode must be protected from mechanical
damage and from corrosion. Accidental disconnection must be avoided by fixing a
permanent label as shown in {Fig 5.17} which reads:

Fig 5.17 Connection of earthing conductor to earth electrode

5.6.1 - What is protective multiple earthing?

If a continuous metallic earth conductor exists from the star point of the supply
transformer to the earthing terminal of the installation, it will run throughout in parallel
with the installation neutral, which will be at the same potential. It therefore seems
logical that one of these conductors should be removed, with that remaining acting as a
combined protective and neutral conductor (PEN). When this is done, we have a TN-C-S
installation {5.2.4}. The combined earth and neutral system will apply only to the supply,
and not to the installation.

Because of possible dangers with the system which will be explained in the following
sub-sections, PME can be installed by the Electricity Supply Company only after the
supply system and the installations it feeds have complied with certain requirements.
These special needs will be outlined in {5.6.4}.
The great virtue of the PME system is that neutral is bonded to earth so that a phase to
earth fault is automatically a phase to neutral fault. The earth-fault loop impedance will
then be low, resulting in a high value of fault current which will operate the protective
device quickly. It must be stressed that the neutral and earth conductors are kept quite
separate within the installation: the main earthing terminal is bonded to the incoming
combined earth and neutral conductor by the Electricity Supply Company. The difficulty
of ensuring that bonding requirements are met on construction sites means that PME
supplies must not be used. Electricity Supply Regulations forbid the use of PME supplies
to feed caravans and caravan sites.

5.6.2 - Increased fire risk

As with other systems of earth-fault protection, PME does not prevent a fault occurring,
but will ensure that the fault protection device operates quickly when that fault appears.
For example, if a fault of 2 Ohms resistance occurs in a 240 V circuit protected by a 20 A
semi-enclosed fuse in a system with an earth-fault loop impedance of 6 Ohms, the fault
current will be 240/(2 + 6) A = 240/8 A = 30 A. The fuse would not blow unless the
circuit were already loaded, when load current would add to fault current. If the circuit
were fully loaded with a load current of 20 A, total current would be 50 A and the fuse
would blow after about 18 s. During this time, the power produced in the fault would be:

P = I²R = 30²x6 = 5400W or 5.4kW

This could easily start a fire. If, however, the earth-fault loop impedance were I Ohm,
current would be 80 A and the fuse would blow in about 1.6 s and limit the energy in the
fault circuit.

5.6.3 - Broken neutral conductor

The neutral of a supply is often common to a large number of installations. In the

(unlikely) event of a broken neutral, all the consumers on the load side of the break could
have a combined neutral and earth potential of the same level as the phase system (240 V
to earth). This situation could be very dangerous, because all earthed metalwork would be
at 240 V above the potential of the general mass of earth.

To prevent such an event, the Electricity Supply Company connects its combined neutral
and earth conductor to earth electrodes at frequent intervals along its run. Whilst this does
not entirely remove the danger, it is much reduced. For example, the assumed earth
resistance values of {Fig 5.18} show that the maximum possible potential to earth in this
case would be 96 V. In practice, much lower resistance values for the earth connections
will reduce this voltage. The Electricity Supply Company goes to very great lengths to
ensure the integrity of its neutral conductor.
 Fig 5.18 Danger due to broken neutral in a PME system

5.6.4 - Special requirements PME-fed installations

An installation connected to a protective multiple earth supply is subject to special

requirements concerning the size of earthing and bonding leads, which are generally
larger in cross-section than those for installations fed by supplies with other types of
earthing. Full discussions with the Electricity Supply Company are necessary before
commencing such an installation to ensure that their needs will be satisfied. The cross-
sectional area of the equipotential bonding conductor is related to that of the neutral
conductor as shown in {Table 5.9}

Table 5.9 - Minimum cross-sectional area of main equipotential

----------------- bonding conductor for PME-fed installations
Neutral conductor c.s.a. Main equipotential bonding conductor c.s.a
(mm²) (mm²)
35 or less 10
Over 35 and up to 50 16
Over 50 and up to 95 25

Danger can arise when the non-current carrying metalwork of an installation is connected
to the neutral, as is the case with a PME-fed system. The earth system is effectively in
parallel with the neutral, and will thus share the normal neutral current. This current will
not only be that drawn by the installation itself, but may also be part of the neutral current
of neighbouring installations.

It follows that the earth system for an installation may carry significant current (of the
order of tens of amperes) even when the main supply to that installation is switched off.
This could clearly cause a hazard if a potentially explosive part of an installation, such as
a petrol storage tank, were the effective earth electrode for part of the neutral current of a
number of installations. For this reason, the Health and Safety Executive has banned the
use of PME in supplies for petrol filling stations. Such installations must be fed from TN-
S supply systems (RSE booklet RS(C)41
- 'Petrol Filling Stations: Construction and Operation').