---

• thermal effects due to resistive heating;

• mechanical effects linked with the magnetic interaction where the lightning current is shared by conductors positioned in the vicinity of one another or when the current changes direction (bends or connections between conductors positioned at a given angle with respect to one another).

In most cases, these two effects act independently from each other and separate laboratory tests can be carried out to check each effect from the other. This approach can be adopted in all cases in which the heating developed by the lightning current flow does not modify substantially the mechanical characteristics.

D.5.3.1 Resistive heating

Calculations and measurements relating to the heating of conductors of different cross­sections and materials due to lightning current flowing along a conductor have been published by several authors. The main results in terms of plots and formulae are summarized in D.4.1.1. No laboratory test is therefore necessary, in general, to check the behaviour of a conductor with respect to temperature rise.

In all cases for which a laboratory test is required, the following considerations shall be taken into account:

the main test parameters to be considered are the specific energy and the impulse current duration;

the specific energy governs the temperature rise due to the Joule heating caused by the flow of the lightning current. Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes;

the impulse current duration has a decisive influence on the heat exchange process with respect to the ambient conditions surrounding the considered conductor. In most cases the duration of the impulse current is so short that the heating process can be considered to be adiabatic.

D.5.3.2 Mechanical effects

As discussed in D.4.2.1, mechanical interactions are developed between conductors carrying lightning current. The force is proportional to the product of the currents flowing in the conductors (or to the square of the current if a single bent conductor is considered) and is linked with the inverse of the distance between the conductors.

The usual situation in which a visible effect can occur is when a conductor forms a loop or is bent. When such a conductor carries the lightning current, it will be subjected to a mechanical force which tries to extend the loop and to straighten the corner and thus to bend it outward. The magnitude of this force is proportional to the square of the current amplitude. A clear distinction should be made, however, between the electrodynamic force, which is proportional to the square of the current amplitude, and the corresponding stress dependent on the elastic characteristics of the mechanical LPS structure. For LPS structures of relatively low natural frequencies, the stress developed within the LPS structure would be considerably lower than the electrodynamic force. In this case, no laboratory test is necessary to check the mechanical behaviour of a conductor bent at a right-angle as long as the cross-sectional areas of the present standard requirements are fulfilled.

In all cases for which a laboratory test is required (especially for soft materials), the following considerations should be taken into account. Three parameters of the first return stroke are to be considered: the duration, the specific energy of the impulse current and, in the case of rigid systems, the amplitude of the current.

The duration of the impulse current, compared with the period of the natural mechanical oscillation of the LPS structure, governs the type of mechanical response of the system in terms of displacement:

• If the duration of the impulse is much shorter than the period of natural mechanical oscillation of the LPS structure (normal case for LPS structures stressed by lightning impulses), the mass and elasticity of the system prevents it from being displaced appreciably and the relevant mechanical stress is essentially related to the specific energy of the current impulse. The peak value of the impulse current has a limited effect.

• If the duration of the impulse is comparable with or higher than the period of natural mechanical oscillation of the structure, the displacement of the system is more sensitive to the shape of the applied stress. In this case, the peak value of the current impulse and its specific energy needs to be reproduced during the test.

The specific energy of the impulse current governs the stress causing the elastic and plastic deformation of the LPS structure. Numerical values to be considered are those relevant to the first stroke.The maximum values of the impulse current govern the length of the maximum displacement of the LPS structure, in case of rigid systems having high natural oscillation frequencies. Numerical values to be considered are those relevant to the first stroke.

D.5.3.3 Connecting components

Connecting components between adjacent conductors of an LPS are possible points of mechanical and thermal weakness where very high stresses occur.

In the case of a connector placed in such a manner as to make the conductor follow a right angle, the main effects of the stresses are linked with mechanical forces which tend to straighten the conductor set and overcome the friction forces between the connecting component and the conductors, thus pulling the connection apart. The development of arcs at the points of contact of the different parts is possible. Moreover, the heating effect caused by the concentration of current over small contact surfaces has a notable effect.

Laboratory tests have shown that it is difficult to separate each effect from the others as a complex synergism takes place. Mechanical strength is affected by local melting of the area of contact. Relative displacements between parts of the connection components promote the development of arcs and the consequential intense heat generation.

In the absence of a valid model, laboratory tests should be conducted in such a way as to represent as closely as possible the appropriate parameters of the lightning current in the most critical situation, i.e. the appropriate parameters of the lightning current shall be applied by means of a single electrical test.

Three parameters should be considered in this case: the peak value, the specific energy and the duration of the impulse current.

The maximum values of the impulse current govern the maximum force, or, if and after the electrodynamic pulling force exceeds the friction force, the length of the maximum displacement of the LPS structure. Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes.

The specific energy of the current impulse governs the heating at contact surfaces where the current is concentrated over small areas. Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes.

The duration of the impulse current governs the maximum displacement of the structure after friction forces are exceeded and has an important role in the heat transfer phenomena into the material.

D.5.3.4 Earth-termination

The real problems with earth-termination electrodes are linked with chemical corrosion and mechanical damage caused by forces other than electrodynamic forces. In practical cases, erosion of the earth electrode at the arc root is of minor importance. It is, however, to be considered that, contrary to air-terminations, a typical LPS has several earth-terminations. The lightning current will be shared between several earthing electrodes, thus causing less important effects at the arc root. Two main test parameters should be considered in this case:

• the charge governs the energy input at the arc root. In particular, the contribution of the first stroke can be neglected since long duration strokes appear to be the most severe for this component;

• the duration of the current impulse has an important role in the heat transfer phenomena into the material. The duration of the current impulses applied during the testing should be comparable to those of long duration strokes (0,5 s to 1s).

D.6 Surge protective device (SPD)

D.6.1 General

The effects of the stress on an SPD caused by lightning depend on the type of SPD considered, with particular reference to the presence or absence of a gap.

D.6.2 SPD containing spark gaps

Effects on spark gaps caused by lightning can be divided into two major categories:

• the erosion of the gap electrodes by heating, melting and vaporizing of material;

• the mechanical stress caused by the shock wave of the discharge.

It is extremely difficult to investigate separately these effects, as both are linked with the main lightning current parameters by means of complex relationships.

For spark gaps, laboratory tests shall be conducted in such a way as to represent as closely as possible the appropriate parameters of the lightning current in the most critical situation, i.e. all the appropriate parameters of the lightning current shall be applied by means of a single electrical stress.

Five parameters shall be considered in this case: the peak value, the charge, the duration, the specific energy and the rate of rise of the impulse current.

The current peak value governs the severity of the shockwave. Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes.

The charge governs the energy input in the arc. The energy in the arc will heat up, melt and possibly vaporize part of the electrode material at the attachment point of the arc. Numerical values to be considered are those relevant to the whole lightning flash. However, the charge of the long duration current can be neglected in many cases, depending on the configuration of the power supply system (TN, TT or IT).

The duration of the impulse current governs the heat transfer phenomena into the mass of the electrode and the resulting propagation of the melt front.

The specific energy of the current impulse governs the self-magnetic compression of the arc and the physics of the electrode plasma jets developed at the interface between the electrode surface and the arc (which can blow out a significant amount of molten material). Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes.

NOTE For spark gaps used on power supply systems, the possible power frequency follow current amplitude constitutes an important stress factor, which must be taken into consideration.

D.6.3 SPD containing metal-oxide varistors

Stress to metal-oxide varistors caused by lightning can be divided into two main categories: overload and flashover. Each category is characterized by failure modes generated by different phenomena and governed by different parameters. The failure of a metal-oxide SPD is linked with its weakest characteristics and therefore it is unlikely that synergism between different fatal stresses can occur. It appears, therefore, to be acceptable to carry out separate tests to check the behaviour under each failure mode condition.

Overloads are caused by an amount of absorbed energy exceeding the capabilities of the device. The excessive energy considered here is related to the lightning stress itself.

However, for SPDs installed on power supply systems, the follow current injected in the device by the power system immediately after the cessation of the lightning current flow can also play an important role in the fatal damage of the SPD. Finally, an SPD can be fatally damaged by thermal instability under the applied voltage related to the negative temperature coefficient of the volt-ampere characteristics of the resistors. For the overload simulation of metal-oxide varistors, one main parameter is to be considered: the charge.

The charge governs the energy input into the metal-oxide resistor block, considering as a constant the residual voltage of the metal-oxide resistor block. Numerical values to be considered are those relevant to the lightning flash.

Flashovers and cracking are caused by the amplitude of current impulses exceeding the capabilities of the resistors. This failure mode is generally evidenced by an external flashover along the collar, sometimes penetrating into the resistor block causing a crack or a hole perpendicular to the collar. The failure is mainly linked with a dielectric collapse of the collar of the resistor block.

For the simulation of this lightning phenomenon, two main parameters should be considered: the maximum value and the duration of the impulse current.

The maximum value of the impulse current determines, through the corresponding level of residual voltage, whether the maximum dielectric strength on the resistor collar is exceeded. Numerical values to be considered are those relevant to the first stroke. Conservative data are obtained by considering positive strokes.

The duration of the impulse current governs the duration of application of the dielectric stress on the resistor collar.

D.7 Summary of the test parameters to be adopted in testing LPS components

Table D.1 summarizes the most critical aspects of each LPS component during the performance of its function and gives the parameters of the lightning current to be reproduced in laboratory tests.

The numerical values given in Table D.1 are relevant to the lightning parameters of importance at the point of strike.

Test values should be calculated considering the current sharing which can be expressed by means of the current sharing factor, as discussed in Clause D.3.

The numerical values of the parameters to be used during the tests can therefore be calculated on the base of the data given in Table D.1, applying the reduction factors linked with current sharing, as expressed by the formulae reported in Clause D.3

.Annex Е
(informative)

Surges due to lightning at different installation points

E.1 Overview

For dimensioning of conductors, SPDs and apparatus, the threat due to surges at the particular installation point of these components should be determined. Surges can arise from (partial) lightning currents and from induction effects into installation loops. The threat due to these surges must be lower than the withstand levels of the components used (defined by adequate tests as necessary).

E.2 Surges due to flashes to the structure (source of damage S1)

E.2.1 Surges flowing through external conductive parts and lines connected to the structure

When conducted to earth, the lightning current is divided between the earth-termination system, the external conductive parts and the lines, directly or via SPDs connected to them.

If /_F = k_e x / (E.1)

is the part of the lightning current relevant to each external conductive part or line, then the current sharing factor k_e depends on:

• the number of parallel paths;

• their conventional earthing impedance for underground parts, or their earth resistance, where overhead parts connect to underground, for overhead parts;

• the conventional earthing impedance of the earth-termination system.

• for underground installation k_e= - -— (E.2)

Z_y+Zx(nf +n₂x —L)

^Z2

• for overhead installation k_e= - — (E.3)

Z₂ + Zx(n₂+n_yx-^-)

where

Z is the conventional earthing impedance of the earth-termination system;

Z₁ is the conventional earthing impedance of the external parts or lines (Table E.1) running underground;

Z₂ is the earth resistance of the earthing arrangement connecting the overhead line to ground. If the earth resistance of the earthing point is not known, the value of Z_y shown in Table E.1 may be used (where the resistivity is relevant to the earthing point).

NOTE 1 This value is assumed in the above formula to be the same for each earthing point. If this is not the case, more complex equations need to be used.

n₁ is the overall number of external parts or lines running underground;

n₂ is the overall number of external parts or lines running overhead;

I is the lightning current relevant to the lightning protection level (LPL) considered.

Assuming as a first approximation that one half of the lightning current flows in the earth­termination system and that Z₂ = Z₁₍ the value of k_e may be evaluated for an external conductive part or line by:

k

(E.4)

_e- 0,5 / (n₁+n₂)

If entering lines (e.g. electrical and telecommunication lines) are unshielded or not routed in metal conduit, each of the n' conductors of the line carries an equal part of the lightning current

к’_е=к_е/п' (E.5)

n’ being the total number of conductors. For shielded lines bonded at the entrance, the values of current sharing factor k’_e for each of the n' conductors of a shielded line are given by:

k'_e= k_ex R_s/(п'x R_s+R_c) (E.6)

where R_s is the ohmic resistance per unit length of shield; R_c is the ohmic resistance per unit length of inner conductor.

NOTE 2 This formula may underestimate the role of the shield in diverting lightning current due to mutual inductance between core and shield.

Table E.1 - Conventional earthing impedance values Z and Z₁
according to the resistivity of the soil

E.2.2 Factors influencing the sharing of the lightning current in power lines

For detailed calculations, several factors can influence the amplitude and the shape of such surges:

• the cable length can influence current sharing and shape characteristics due to the L/R ratio;

• different impedances of neutral and phase conductors can influence current sharing among line conductors;

NOTE 1 For example, if the neutral (N) conductor has multiple earths, the lower impedance of N compared with phase conductors L_1f L₂, and L₃ could result in 50 % of the current flowing through the N conductor with the remaining 50 % being shared by the other 3 phase conductors (17 % each). If N, L-,, L₂, and L₃ have the same impedance, each conductor will carry approximately 25 % of the current.

• different transformer impedances can influence current sharing (this effect is negligible, if the transformer is protected by SPDs bypassing its impedance);

• the relation between the conventional earthing resistances of the transformer and the items on the load side can influence current sharing (the lower the transformer impedance, the higher is the surge current flowing into the low voltage system);

• parallel consumers cause a reduction of the effective impedance of the low voltage system; this may increase the partial lightning current flowing into this system.

NOTE 2 Refer to Annex D of IEC 62305-4:2010 for more information.

E.3 Surges relevant to lines connected to the structure

E.3.1 Surges due to flashes to lines (source of damage S3)

For direct lightning flashes to connected lines, partitioning of the lightning current in both directions of the line and the breakdown of insulation should be taken into account.

The selection of the /_imp value can be based on values given in Tables E.2 and E.3 for low- voltage systems and Table E.3 for telecommunication systems where the preferred values of /_jmp are associated with the lightning protection level (LPL).

Table E.2 - Expected surge overcurrents due to lightning flashes
on low-voltage systems

NOTE All values refer to each line conductor.

^a Loop conductors routing and distance from inducing current affect the values of expected surge overcurrents.

Values in Table E.2 refer to short-circuited, unshielded loop conductors with different routing in large buildings (loop area in the order of 50 m , width = 5 m), 1 m apart from the structure wall, inside an unshielded structure or building with LPS (k_c= 0,5). For other loop and structure characteristics, values should be multiplied by factors K_S1, K_S2, K_S3 (see Clause B.4 of IEC 62305-2:2010).