0100 Electrical Installation Guide Guide En

contents

A

Contents

A. contents A1B. general - installed power A1C. HV/LV distribution substations A2D. low-voltage service connections A3E. power factor improvement and harmonic filtering A3F. distribution within a low-voltage installation A4G. protection against electric shocks A5H. the protection of circuits and the switchgear A7H1. the protection of circuits A7H2. the switchgear A9J. particular supply sources and loads A10L. domestic and similar premises and special locations A12

<< Back

contents - A1

A

contents

A. contents A1

B. general - installed power

1. methodology B1

2. rules and statutory regulations B3

2.1 definition of voltage ranges B3table B1standard voltages between 100 V and 1000 V (IEC 38-1983) B3table B2standard voltages above 1 kV and not exceeding 35 kV (IEC 38-1983) B3

2.2 regulations B4

2.3 standards B4

2.4 quality and safety of an electrical installation B5

2.5 initial testing of an installation B6

2.6 periodic check-testing of an installation B6table B3frequency of check-tests commonly recommended for an electrical installation B6

2.7 conformity (with standards and specifications) of equipmentused in the installation B7

3. motor, heating and lighting loads B8

3.1 induction motors B8table B4power and current values for typical induction motors B9

3.2 direct-current motors B10table B6progressive starters with voltage ramp B10table B7progressive starters with current limitation B10

3.3 resistive-type heating appliances and incandescent lamps(conventional or halogen) B11table B8current demands of resistive heating and incandescent lighting (conventional or halogen)appliances B11

3.4 fluorescent lamps and related equipment B11table B10current demands and power consumption of commonly-dimensioned fluorescentlighting tubes (at 220 V/240 V - 50 Hz) B12table B11current demands and power consumption of compact fluorescent lamps(at 220 V/240 V - 50 Hz) B12

3.5 discharge lamps B13table B12current demands of discharge lamps B13

4. power loading of an installation B14

4.1 installed power (kW) B14

4.2 installed apparent power (kVA) B15table B13estimation of installed apparent power B15

4.3 estimation of actual maximum kVA demand B16table B14simultaneity factors in an apartment block B16table B16factor of simultaneity for distribution boards (IEC 439) B17table B17factor of simultaneity according to circuit function B17

4.4 example of application of factors ku and ks B17table B18an example in estimating the maximum predicted loading of an installation(the factor values used are for demonstration purposes only) B17

A2 - contents

A

contents (continued)

B. general - installed power (continued)

4. power loading of an installation (continued)

4.5 diversity factor B18

4.6 choice of transformer rating B18table B19IEC-standardized kVA ratings of HV/LV 3-phase distribution transformersand corresponding nominal full-load current values B18

4.7 choice of power-supply sources B19

C. HV/LV distribution substations

1. supply of power at high voltage C1

1.1 power-supply characteristics of high voltage distribution networks C1table C1relating nominal system voltages with corresponding rated system voltages(r.m.s. values) C2table C2switchgear rated insulation levels C3table C3Atransformers rated insulation levels in series I (based on current practice other thanin the United States of America and some other countries) C3table C3Btransformers rated insulation levels in series II (based on current practice in the UnitedStates of America and some other countries) C4table C4standard short-circuit current-breaking ratings extracted from table X IEC 56 C4

1.2 different HV service connections C11

1.3 some operational aspects of HV distribution networks C13

2. consumers HV substations C15

2.1 procedures for the establishment of a new substation C15

3. substation protection schemes C17

3.1 protection against electric shocks and overvoltages C17

3.2 electrical protection C22table C18power limits of transformers with a maximum primary current not exceeding 45 A C25table C19rated current (A) of HV fuses for transformer protection according to IEC 282-1 C26table C203-phase short-circuit currents of typical distribution transformers C27

3.3 protection against thermal effects C31

3.4 interlocks and conditioned manœuvres C31

4. the consumer substation with LV metering C34

4.1 general C34

4.2 choice of panels C36table C27standard short-circuit MVA and current ratings at different levels of nominal voltage C37

4.3 choice of HV switchgear panel for a transformer circuit C38

4.4 choice of HV/LV transformer C38table C31categories of dielectric fluids C41table C32safety measures recommended in electrical installations using dielectric liquidsof classes 01, K1, K2 or K3 C42

5. a consumer substation with HV metering C44

5.1 general C44

5.2 choice of panels C46

5.3 parallel operation of transformers C48

6. constitution of HV/LV distribution substations C49

6.1 different types of substation C49

6.2 indoor substations equipped with metal-enclosed switchgear C49

6.3 outdoor substations C52

contents - A3

A

D. low-voltage service connections

1. low-voltage public distribution networks D1

1.1 low-voltage consumers D1table D1survey of electricity supplies in various countries around the world. D1table D2 D6

1.2 LV distribution networks D7

1.3 the consumer-service connection D10

1.4 quality of supply voltage D13

2. tariffs and metering D14

E. power factor improvement and harmonic filtering

1. power factor improvement E1

1.1 the nature of reactive energy E1

1.2 plant and appliances requiring reactive current E2

1.3 the power factor E2

1.4 tan ϕ E3

1.5 practical measurement of power factor E4

1.6 practical values of power factor E4table E5example in the calculation of active and reactive power E4table E7values of cos ϕ and tan ϕ for commonly-used plant and equipment E4

2. why improve the power factor? E5

2.1 reduction in the cost of electricity E5

2.2 technical/economic optimization E5table E8multiplying factor for cable size as a function of cos ϕ E5

3. how to improve the power factor E6

3.1 theoretical principles E6

3.2 by using what equipment? E7

3.3 the choice between a fixed or automatically-regulated bankof capacitors E8

4. where to install correction capacitors E9

4.1 global compensation E9

4.2 compensation by sector E9

4.3 individual compensation E10

5. how to decide the optimum level of compensation E11

5.1 general method E11

5.2 simplified method E11table E17kvar to be installed per kW of load, to improve the power factor of an installation E12

5.3 method based on the avoidance of tariff penalties E13

5.4 method based on reduction of declared maximum apparentpower (kVA) E13

6. compensation at the terminals of a transformer E14

6.1 compensation to increase the available active power output E14table E20active-power capability of fully-loaded transformers, when supplying loads at differentvalues of power factor E14

6.2 compensation of reactive energy absorbed by the transformer E15table E24reactive power consumption of distribution transformers with 20 kV primary windings E16

A4 - contents

A


E. power factor improvement and harmonic filtering (continued)

7. compensation at the terminals of an induction motor E17

7.1 connection of a capacitor bank and protection settings E17table E26reduction factor for overcurrent protection after compensation E17

7.2 how self-excitation of an induction motor can be avoided E18table E28maximum kvar of P.F. correction applicable to motor terminals without riskof self-excitation E19

8. example of an installation before and after

power-factor correction E20

9. the effect of harmonics on the rating of a capacitor

bank E21

9.1 problems arising from power-system harmonics E21

9.2 possible solutions E21

9.3 choosing the optimum solution E22table E30choice of solutions for limiting harmonics associated with a LV capacitor bank E22

9.4 possible effects of power-factor-correction capacitorson the power-supply system E23

10. implementation of capacitor banks E24

10.1 capacitor elements E24

10.2 choice of protection, control devices, and connecting cables E25

F. distribution within a low-voltage installation

1. general F1

1.1 the principal schemes of LV distribution F1

1.2 the main LV distribution board F4

1.3 transition from IT to TN F4

2. essential services standby supplies F5

2.1 continuity of electric-power supply F5

2.2 quality of electric-power supply F6table F10assumed levels of transient overvoltage possible at different points of a typicalinstallation F8table F12typical levels of impulse withstand voltage of industrial circuit breakers labelledUimp = 8 kV F8table F18compatibility levels for installation materials F13

3. safety and emergency-services installations,

and standby power supplies F15

3.1 safety installations F15

3.2 standby reserve-power supplies F15

3.3 choice and characteristics of reserve-power supplies F16table F21table showing the choice of reserve-power supply types according to applicationrequirements and acceptable supply-interruption times F16

3.4 choice and characteristics of different sources F17table F22table of characteristics of different sources F17

3.5 local generating sets F18

contents - A5

A

4. earthing schemes F19

4.1 earthing connections F19table F25list of exposed-conductive-parts and extraneous-conductive-parts F204.2 definition of standardized earthing schemes F21

4.3 earthing schemes characteristics F23

4.4.1 choice criteria F294.4.2 comparison for each criterion F30

4.5 choice of earthing method - implementation F31

4.6 installation and measurements of earth electrodes F32table F47resistivity (Ω-m) for different kinds of terrain F33table F48mean values of resistivity (Ω-m) for an approximate estimation of an earth-electroderesistance with respect to zero-potential earth F33

5. distribution boards F36

5.1 types of distribution board F36

5.2 the technologies of functional distribution boards F37

5.3 standards F38

5.4 centralized control F38

6. distributors F39

6.1 description and choice F39

6.2 conduits, conductors and cables F41table F60selection of wiring systems F41table F61erection of wiring systems F41table F62some examples of installation methods F43table F63designation code for conduits according to the most recent IEC publications F44table F64designation of conductors and cables according to CENELEC code for harmonizedcables F45table F66commonly used conductors and cables F46

7. external influences F47

7.1 classification F47table F67concise list of important external influences (taken from Appendix A of IEC 364-3) F487.2 protection by enclosures: IP code F49

G. protection against electric shocks

1. general G1

1.1 electric shock G1

1.2 direct and indirect contact G1

2. protection against direct contact G2

2.1 measures of protection against direct contact G22.2 additional measure of protection against direct contact G3

3. protection against indirect contact G4

3.1 measure of protection by automatic disconnection of the supply G4table G8maximum safe duration of the assumed values of touch voltage in conditions whereUL = 50 V G4table G9maximum safe duration of the assumed values of touch voltage in conditions whereUL = 25 V G4

3.2 automatic disconnection for a TT-earthed installation G5table G11maximum operating times of RCCBs (IEC 1008) G6

3.3 automatic disconnection for a TN-earthed installation G6table G13maximum disconnection times specified for TN earthing schemes (IEC 364-4-41) G7

A6 - contents

A


G. protection against electric shocks (continued)

3. protection against indirect contact (continued)

3.4 automatic disconnection on a second earth fault in an IT-earthedsystem G8table G18maximum disconnection times specified for an IT-earthed installation (IEC 364-4-41) G9

3.5 measures of protection against direct or indirect contactwithout circuit disconnection G10

4. implementation of the TT system G13

4.1 protective measures G13table G26the upper limit of resistance for an installation earthing electrode which must not beexceeded, for given sensitivity levels of RCDs at UL voltage limits of 50 V and 25 V G13

4.2 types of RCD G14

4.3 coordination of differential protective devices G15

5. implementation of the TN system G18

5.1 preliminary conditions G18

5.2 protection against indirect contact G18table G42correction factor to apply to the lengths given in tables G43 to G46 for TN systems G20table G43maximum circuit lengths for different sizes of conductor andinstantaneous-tripping-current settings for general-purpose circuit breakers G20table G44maximum circuit lengths for different sizes of conductor and rated currents for type Bcircuit breakers G20table G45maximum circuit lengths for different conductor sizes and for rated currents of circuitbreakers of type C G21table G46maximum circuit lengths for different conductor sizes and for rated currents of circuitbreakers of type D or MA Merlin Gerin G21

5.3 high-sensitivity RCDs G22

5.4 protection in high fire-risk locations G22

5.5 when the fault-current-loop impedance is particularly high G23

6. implementation of the IT system G24

6.1 preliminary conditions G24table G53essential functions in IT schemes G24

6.2 protection against indirect contact G25table G59correction factors, for IT-earthed systems, to apply to the circuit lengths givenin tables G43 to G46 G28

6.3 high-sensitivity RCDs G29

6.4 in areas of high fire-risk G29

6.5 when the fault-current-loop impedance is particularly high G30

7. residual current differential devices (RCDs) G31

7.1 description G31

7.2 application of RCDs G31table G70electromagnetic compatibility withstand-level tests for RCDs G32table G72means of reducing the ratio I∆n/lph (max.) G33

7.3 choice of characteristics of a residual-current circuit breaker(RCCB - IEC 1008) G34table G74typical manufacturers coordination table for RCCBs, circuit breakers, and fuses G34

contents - A7

A

H. the protection of circuits and the switchgear

H1. the protection of circuits

1. general H1-1

1.1 methodology and definitions H1-1table H1-1logigram for the selection of cable size and protective-device rating for a given circuit H1-1

1.2 overcurrent protection principles H1-3

1.3 practical values for a protection scheme H1-4

1.4 location of protective devices H1-5table H1-7general rules and exceptions concerning the location of protective devices H1-5

1.5 cables in parallel H1-5

1.6 worked example of cable calculations H1-6table H1-9calculations carried out with ECODIAL software (Merlin Gerin) H1-8table H1-10example of short-circuit current evaluation H1-9

2. practical method for determining the smallest

allowable cross-sectional-area of circuit conductors H1-10

2.1 general H1-10table H1-11logigram for the determination of minimum conductor size for a circuit H1-10

2.2 determination of conductor size for unburied circuits H1-10table H1-12code-letter reference, depending on type of conductor and method of installation H1-10table H1-13factor K1 according to method of circuit installation (for further examples referto IEC 364-5-52 table 52H) H1-11table H1-14correction factor K2 for a group of conductors in a single layer H1-11table H1-15correction factor K3 for ambient temperature other than 30 °C H1-12table H1-17case of an unburied circuit: determination of the minimum cable size (c.s.a.), derivedfrom the code letter; conductor material; insulation material and the fictitious current I'z H1-13

2.3 determination of conductor size for buried circuits H1-14table H1-19correction factor K4 related to the method of installation H1-14table H1-20correction factor K5 for the grouping of several circuits in one layer H1-14table H1-21correction factor K6 for the nature of the soil H1-15table H1-22correction factor K7 for soil temperatures different than 20 °C H1-15table H1-24case of a buried circuit: minimum c.s.a. in terms of type of conductor; type of insulation;and value of fictitious current I'z (I'z = Iz) H1-15 K

3. determination of voltage drop H1-17

3.1 maximum voltage-drop limit H1-17table H1-26maximum voltage-drop limits H1-17

3.2 calculation of voltage drops in steady load conditions H1-18table H1-28voltage-drop formulae H1-18table H1-29phase-to-phase voltage drop ∆U for a circuit, in volts per ampere per km H1-18

A8 - contents

A


H. the protection of circuits and the switchgear (continued)

H1. the protection of circuits (continued)

4. short-circuit current calculations H1-20

4.1 short-circuit current at the secondary terminals of a HV/LVdistribution transformer H1-20table H1-32typical values of Usc for different kVA ratings of transformers with HV windings i 20 kV H1-20table H1-33Isc at the LV terminals of 3-phase HV/LV transformers supplied from a HV systemwith a 3-phase fault level of 500 MVA, or 250 MVA H1-20

4.2 3-phase short-circuit current (Isc) at any point withina LV installation H1-21table H1-36the impedance of the HV network referred to the LV side of the HV/LV transformer H1-21table H1-37resistance, reactance and impedance values for typical distribution transformerswith HV windings i 20 kV H1-22table H1-38recapitulation table of impedances for different parts of a power-supply system H1-23table H1-39example of short-circuit current calculations for a LV installation supplied at 400 V(nominal) from a 1,000 kVA HV/LV transformer H1-23

4.3 Isc at the receiving end of a feeder in terms of the Iscat its sending end H1-23table H1-40Isc at a point downstream, in terms of a known upstream fault-current valueand the length and c.s.a. of the intervening conductors, in a 230/400 V 3-phase system H1-24

4.4 short-circuit current supplied by an alternator or an inverter H1-25

5. particular cases of short-circuit current H1-26

5.1 calculation of minimum levels of short-circuit current H1-26table H1-49maximum circuit lengths in metres for copper conductors (for aluminium, the lengthsmust be multiplied by 0.62) H1-28table H1-50maximum length of copper-conductored circuits in metres protected by B-typecircuit breakers H1-29table H1-51maximum length of copper-conductored circuits in metres protected by C-typecircuit breakers H1-29table H1-52maximum length of copper-conductored circuits in metres protected by D-typecircuit breakers H1-29table H1-53correction factors to apply to lengths obtained from tables H1-49 to H1-52 H1-30

5.2 verification of the withstand capabilities of cablesunder short-circuit conditions H1-31table H1-54value of the constant k2 H1-31table H1-55maximum allowable thermal stress for cables (expressed in amperes2 x seconds x 106) H1-31

6. protective earthing conductors (PE) H1-32

6.1 connection and choice H1-32table H1-59choice of protective conductors (PE) H1-33

6.2 conductor dimensioning H1-33table H1-60minimum c.s.a.'s for PE conductors and earthing conductors(to the installation earth electrode) H1-34table H1-61k factor values for LV PE conductors, commonly used in national standardsand complying with IEC 724 H1-34

6.3 protective conductor between the HV/LV transformerand the main general distribution board (MGDB) H1-35table H1-63c.s.a. of PE conductor between the HV/LV transformer and the MGDB, in terms of transformerratings and fault-clearance times used in France H1-35

6.4 equipotential conductor H1-35

contents - A9

A

7. the neutral conductor H1-36

7.1 dimensioning the neutral conductor H1-36

7.2 protection of the neutral conductor H1-36table H1-65table of protection schemes for neutral conductors in different earthing systems H1-37

H2. the switchgear

1. the basic functions of LV switchgear H2-1table H2-1basic functions of LV switchgear H2-1

1.1 electrical protection H2-1

1.2 isolation H2-1table H2-2peak value of impulse voltage according to normal service voltage of test specimen H2-2

1.3 switchgear control H2-2

2. the switchgear and fusegear H2-4

2.1 elementary switching devices H2-4table H2-7utilization categories of LV a.c. switches according to IEC 947-3 H2-5table H2-8factor "n" used for peak-to-rms value (IEC 947-part 1) H2-5table H2-13zones of fusing and non-fusing for LV types gG and gM class fuses(IEC 269-1 and 269-2-1) H2-7

2.2 combined switchgear elements H2-9

3. choice of switchgear H2-11

3.1 tabulated functional capabilities H2-11table H2-19functions fulfilled by different items of switchgear H2-11

3.2 switchgear selection H2-11

4. circuit breakers H2-12

table H2-20functions performed by a circuit breaker/disconnector H2-12

4.1 standards and descriptions H2-12

4.2 fundamental characteristics of a circuit breaker H2-15table H2-28tripping-current ranges of overload and short-circuit protective devicesfor LV circuit breakers H2-16table H2-31Icu related to power factor (cos ϕ) of fault-current circuit (IEC 947-2) H2-17

4.3 other characteristics of a circuit breaker H2-18table H2-34relation between rated breaking capacity Icu and rated making capacity Icm at differentpower-factor values of short-circuit current, as standardized in IEC 947-2 H2-19

4.4 selection of a circuit breaker H2-20table H2-38examples of tables for the determination of derating/uprating factors to apply to CBswith uncompensated thermal tripping units, according to temperature H2-21table H2-40different tripping units, instantaneous or short-time delayed H2-23table H2-43maximum values of short-circuit current to be interrupted by main and principalcircuit breakers (CBM and CBP respectively), for several transformers in parallel H2-25

4.5 coordination between circuit breakers H2-27table H2-45example of cascading possibilities on a 230/400 V or 240/415 V 3-phase installation H2-28table H2-49summary of methods and components used in order to achieve discriminative tripping H2-29

4.6 discrimination HV/LV in a consumer's substation H2-32

A10 - contents

A


J. particular supply sources and loads

1. protection of circuits supplied by an alternator J1

1.1 an alternator on short-circuit J1

1.2 protection of essential services circuits supplied in emergenciesfrom an alternator J4

1.3 choice of tripping units J5

1.4 methods of approximate calculation J6table J1-7procedure for the calculation of 3-phase short-circuit current J6table J1-8procedure for the calculation of 1-phase to neutral short-circuit current J7

1.5 the protection of standby and mobile a.c. generating sets J9

2. inverters and UPS

(Uninterruptible Power Supply units) J10

2.1 what is an inverter? J10

2.2 types of UPS system J10table J2-4examples of different possibilities and applications of inverters, in decontamination of suppliesand in UPS schemes J11

2.3 standards J11

2.4 choice of a UPS system J12

2.5 UPS systems and their environment J14

2.6 putting into service and technology of UPS systems J15

2.7 earthing schemes J17

2.8 choice of main-supply and circuit cables, and cables for the batteryconnection J20table J2-21voltage drop in % of 324 V d.c. for a copper-cored cable J21table J2-22currents and c.s.a. of copper-cored cables feeding the rectifier, and supplying the loadfor UPS system Maxipac (cable lengths < 100 m) J21table J2-23currents and c.s.a. of copper-cored cables feeding the rectifier, and supplying the loadfor UPS system EPS 2000 (cable lengths < 100 m). Battery cable data are also included J21table J2-24input, output and battery currents for UPS system EPS 5000 (Merlin Gerin) J22

2.9 choice of protection schemes J23

2.10 complementary equipments J24

3. protection of LV/LV transformers J25

3.1 transformer-energizing in-rush current J25

3.2 protection for the supply circuit of a LV/LV transformer J25

3.3 typical electrical characteristics of LV/LV 50 Hz transformers J26table J3-5typical electrical characteristics of LV/LV 50 Hz transformers J26

3.4 protection of transformers with characteristics as tabled in J3-5above, using Merlin Gerin circuit breakers J26table J3-6protection of 3-phase LV/LV transformers with 400 V primary windings J26table J3-7protection of 3-phase LV/LV transformers with 230 V primary windings J27table J3-8protection of 1-phase LV/LV transformers with 400 V primary windings J27table J3-9protection of 1-phase LV/LV transformers with 230 V primary windings J28

contents - A11

A

4. lighting circuits J29

4.1 service continuity J29

4.2 lamps and accessories (luminaires) J30table J4-1analysis of disturbances in fluorescent-lighting circuits J30

4.3 the circuit and its protection J31

4.4 determination of the rated current of the circuit breaker J31table J4-2protective circuit breaker ratings for incandescent lamps and resistive-type heatingcircuits J31table J4-3maximum limit of rated current per outgoing lighting circuit, for high-pressure dischargelamps J32table J4-4current ratings of circuit breakers related to the number of fluorescent luminaires to beprotected J32

4.5 choice of control-switching devices J33table J4-5types of remote control J33

4.6 protection of ELV lighting circuits J34

4.7 supply sources for emergency lighting J35

5. asynchronous motors J36

5.1 protective and control functions required J36table J5-2commonly-used types of LV motor-supply circuits J37

5.2 standards J38

5.3 basic protection schemes: circuit breaker / contactor / thermal relay J38table J5-4utilization categories for contactors (IEC 947-4) J39

5.4 preventive or limitative protection J41

5.5 maximum rating of motors installed for consumers supplied at LV J43table J5-12maximum permitted values of starting current for direct-on-line LV motors (230/400 V) J43table J5-13maximum permitted power ratings for LV direct-on-line-starting motors J43

5.6 reactive-energy compensation (power-factor correction) J43

6. protection of direct-current installations J44

6.1 short-circuit currents J44

6.2 characteristics of faults due to insulation failure, and of protectiveswitchgear J45table J6-4characteristics of protective switchgear according to type of d.c. system earthing J45

6.3 choice of protective device J45table J6-5choice of d.c. circuit breakers manufactured by Merlin Gerin J46

6.4 examples J46

6.5 protection of persons J47

A12 - contents

A


L. domestic and similar premises and special locations

1. domestic and similar premises L1

1.1 general L1

1.2 distribution-board components L2

1.3 protection of persons L4

1.4 circuits L6table L1-9recommended minimum number of lighting and power points in domestic premises L6table L1-11c.s.a. of conductors and current rating of the protective devices in domesticinstallations (the c.s.a. of aluminium conductors are shown in brackets) L7

2. bathrooms and showers L8

2.1 classification of zones L8

2.2 equipotential bonding L10

2.3 requirements prescribed for each zone L10

3. recommendations applicable to special installations

and locations L11

general - installed power

B

General - installed power

1. Methodology B1

2. Rules and statutory regulations2.1 definition of voltage ranges B32.2 regulations B42.3 standards B42.4 quality and safety of an electrical installation B52.5 initial testing of an installation B52.6 periodic check-testing of an installation B62.7 conformity of equipment used in the installation B6

3. Motor, heating and lighting loads3.1 induction motors B83.2 direct-current motors B103.3 resistive-type heating appliances and incandescent lamps B11(conventional or halogen)3.4 fluorescent lamps and related equipment B113.5 discharge lamps B13

4. Power loading of an installation4.1 installed power B144.2 installed apparent power B154.3 estimation of actual maximum kVA demand B164.4 example of application of factors ku and ks B174.5 diversity factor B184.6 choice of transformer rating B184.7 choice of power-supply sources B19

general - installed power - B1

B

1. methodology

the study of an electrical installationby means of this guide requires thereading of the entire text in the orderin which the chapters are presented.

listing of power demandsThe study of a proposed electrical installationnecessitates an adequate understanding ofall governing rules and regulations.A knowledge of the operating modes ofpower-consuming appliances, i.e. "loads"(steady-state demand, starting conditions,non-simultaneous operation, etc.) togetherwith the location and magnitude of each loadshown on a building plan, allow a listing ofpower demands to be compiled. The list willinclude the total power of the loads installedas well as an estimation of the actual loads tobe supplied, as deduced from the operatingmodes.From these data the power required fromthe supply source and (where appropriate)the number of sources necessary for anadequate supply to the installation, arereadily obtained.Local information regarding tariff structures isalso required to permit the best choice ofconnection arrangement to the power-supplynetwork, e.g. at high voltage or low voltage.

corresponding chapterB - general - installed power

service connectionThis connection can be made at:c High Voltage:a consumer-type substation will then have tobe studied, built and equipped. Thissubstation may be an outdoor or indoorinstallation conforming to relevant standardsand regulations (the low-voltage section maybe studied separately if necessary). Meteringat high-voltage or low-voltage is possible inthis casec Low Voltage:the installation will be connected to the localpower network and will (necessarily) bemetered according to LV tariffs.

C - HV/LV distribution substations

D - low-voltage service connections

reactive energyThe compensation of reactive energy withinelectrical installations normally concerns onlypower factor improvement, and is carried outlocally, globally or as a combination of bothmethods.

E - power factor improvement

LV distributionThe whole of the installation distributionnetwork is studied as a complete system.The number and characteristics of standbyemergency-supply sources are defined.Earth-bonding connections and neutral-earthing arrangements are chosen accordingto local regulations, constraints related to thepower-supply, and to the nature of theinstallation loads.The hardware components of distribution,together with distribution boards andcableways, are determined from buildingplans and from the location and grouping ofloads.The kinds of location, and activities practisedin them, can affect their level of resistance toexternal influences.

F - distribution within a low-voltageinstallation

protection against electric shockThe system of earthing (TT, IT or TN) havingbeen previously determined, it remains, inorder to achieve protection of persons againstthe hazards of direct and indirect contact, tochoose an appropriate scheme of protection.

G - protection against electric shock

B2 - general - installed power

B

1. methodology (continued)

circuits and switchgearEach circuit is then studied in detail.From the rated currents of the loads; the levelof short-circuit current; and the type ofprotective device, the cross-sectional area ofcircuit conductors can be determined, takinginto account the nature of the cableways andtheir influence on the current rating ofconductors.Before adopting the conductor size indicatedabove, the following requirements must besatisfied:c the voltage drop complies with the relevantstandard,c motor starting is satisfactory,c protection against electric shock is assured.The short-circuit current Isc is thendetermined, and the Isc thermal and electro-dynamic withstand capability of the circuit ischecked.These calculations may indicate that adifferent conductor size than that originallychosen is necessary.The performance required by the switchgearwill determine its type and characteristics.The use of cascading techniques and thediscriminative operation of fuses and trippingof circuit breakers are examined.

H1 - the protection of circuits

H2 - the switchgear

particular supply sourcesand loadsParticular items of plant and equipment arestudied:c specific sources such as alternators orinverters,c specific loads with special characteristics,such as induction motors, lighting circuits orLV/LV transformers, orc specific systems, such as direct-currentnetworks.

J - particular supply sources and loads

domestic and similar premisesand special locationsCertain premises and locations are subject toparticularly strict regulations: the mostcommon example being domestic dwellings.

L - domestic and similar premises and speciallocations

Ecodial 2.2 softwareEcodial 2.2 software* provides a complete conception and design package for LV installations, inaccordance with IEC standards and recommendations.The following features are included:c construction of one-line diagrams,c calculation of short-circuit currents,c calculation of voltage drops,c optimization of cable sizes,c required ratings of switchgear and fusegear,c discrimination of protective devices,c recommendations for cascading schemes,c verification of the protection of persons,c comprehensive print-out of the foregoing calculated design data.* Ecodial 2.2 is a Merlin Gerin product and is available in French and English versions.


B

three phase, four wire or three wire systems single phase, three wire systemsnominal voltage (V) nominal voltage (V)- 120/240230/400(1) -277/480(2) -400/690(1) -1000 -

2. rules and statutory regulations

Low-voltage installations are governed by anumber of regulatory and advisory texts,which may be classified as follows:c statutory regulations (decrees, factory acts,etc.),c codes of practice, regulations issued byprofessional institutions, job specifications,c national and international standards forinstallations,c national and international standards forproducts.

2.1 definition of voltage ranges

IEC voltage standards and

recommendations

table B1: standard voltages between 100 V and 1000 V (IEC 38-1983).

1) The nominal voltage of existing 220/380 V and 240/415 V systems shall evolve towards the recommended value of230/400 V. The transition period should be as short as possible, and should not exceed 20 years after the issue of this IECpublication. During this period, as a first step, the electricity supply authorities of countries having 220/380 V systems shouldbring the voltage within the range 230/400 V +6% -10% and those of countries having 240/415 V systems should bring thevoltage within the range 230/400 V +10% -6%. At the end of this transition period the tolerance of 230/400 V ±10% should havebeen achieved; after this the reduction of this range will be considered. All the above considerations apply also to the present380/660 V value with respect to the recommended value 400/690 V.2) Not to be utilized together with 230/400 V or 400/690 V.

50 Hz and 60 Hz systems 60 Hz systemsseries I series II (North American practice)highest voltage nominal system highest voltage nominal systemfor equipment (kV) voltage (kV) for equipment (kV) voltage (kV)3.6(1) 3.3(1) 3((1) 4.40(1) 4.16(1)

7.2(1) 6.6(1) 6(1) - -12 11 10 - -- - - 13.2(2) 12.47(2)

- - - 13.97(2) 13.2(2)

- - - 14.52(1) 13.8(1)

(17.5) - (15) - -24 22 20 - -- - - 26.4(2) 24.94(2)

36(3) 33(3) - - -- - - 36.5(2) 34.5(2)

40.5(3) - 35(3) - -

table B2: standard voltages above 1 kV and not exceeding 35 kV (IEC 38-1983).

* These systems are generally three-wire systems unless otherwise indicated. The values indicated are voltages betweenphases.The values indicated in parentheses should be considered as non-preferred values. It is recommended that these values shouldnot be used for new systems to be constructed in future.1) These values should not be used for public distribution systems.2) These systems are generally four-wire systems.3) The unification of these values is under consideration.


B

2. rules and statutory regulations (continued)

2.2 regulations

In most countries, electrical installations shallcomply with more than one set of regulations,issued by National Authorities or byrecognised private bodies. It is essential totake into account these local constraintsbefore starting the design.

2.3 standards

This Guide is based on relevant IECstandards, in particular IEC 364. IEC 364 hasbeen established by medical and engineeringexperts of all countries in the worldcomparing their experience at an internationallevel. Currently, the safety principles ofIEC 364 and 479-1 are the fundamentals ofmost electrical standards in the world.

IEC - 38 Standard voltagesIEC - 56 High-voltage alternating-current circuit breakersIEC - 76-2 Power transformer - Part 2: Temperature riseIEC - 76-3 Power transformer - Part 3: Insulation levels and dielectric testsIEC - 129 Alternating current disconnectors and earthing switchesIEC - 146 General requirements and line commutated convertersIEC - 146-4 General requirements and line commutated converters - Part 4: Method

of specifying the performance and test requirements of uninterruptible powersystems

IEC - 265-1 High-voltage switches - Part 1: High-voltage switches for rated voltages above1 kV and less than 52 kV

IEC - 269-1 Low-voltage fuses - Part 1: General requirementsIEC - 269-3 Low-voltage fuses - Part 3: Supplementary requirements for fuses for use by

unskilled persons (fuses mainly for household and similar applications)IEC - 282-1 High-voltage fuses - Part 1: Current limiting fusesIEC - 287 Calculation of the continuous current rating of cables (100% load factor)IEC - 298 AC metal-enclosed switchgear and controlgear for rated voltages above 1kV

and up to and including 52 kVIEC - 364 Electrical installations of buildingsIEC - 364-3 Electrical installations of buildings - Part 3: Assessment of general

characteristicsIEC - 364-4-41 Electrical installations of buildings - Part 4: Protection of safety - Section 41:

Protection against electrical shockIEC - 364-4-42 Electrical installations of buildings - Part 4: Protection of safety - Section 42:

Protection against thermal effectsIEC - 364-4-43 Electrical installations of buildings - Part 4: Protection of safety - Section 43:

Protection against overcurrentIEC - 364-4-47 Electrical installations of buildings - Part 4: Application of protective measures

for safety - Section 47: Measures of protection against electrical shockIEC - 364-5-51 Electrical installations of buildings - Part 5: Selection and erection of electrical

equipment - Section 51: Common rulesIEC - 364-5-52 Electrical installations of buildings - Part 5: Selection and erection of electrical

equipment - Section 52: Wiring systemsIEC - 364-5-53 Electrical installations of buildings - Part 5: Selection and erection of electrical

equipment - Section 53: Switchgear and controlgearIEC - 364-6 Electrical installations of buildings - Part 6: VerificationIEC - 364-7-701 Electrical installations of buildings - Part 7: Requirements for special

installations or locations - Section 701: Electrical installations in bathroomsIEC - 364-7-706 Electrical installations of buildings - Part 7: Requirements for special

installations or locations - Section 706: Restrictive conductive locationsIEC - 364-7-710 Electrical installations of buildings - Part 7: Requirements for special

installations or locations - Section 710: Installation in exhibitions, shows, standsand funfairs

IEC - 420 High-voltage alternating current switch-fuse combinationsIEC - 439-1 Low-voltage switchgear and controlgear assemblies - Part 1: Types-tested

and partially type-tested assembliesIEC - 439-2 Low-voltage switchgear and controlgear assemblies - Part 2: Particular

requirements for busbar trunking systems (busways)IEC - 439-3 Low-voltage switchgear and controlgear assemblies - Part 3: Particular

requirements for low-voltage switchgear and controlgear assemblies intendedto be installed in places where unskilled persons have access for their use -Distribution boards

IEC - 446 Identification of conductors by colours or numeralsIEC - 479-1 Effects of current on human beings and livestock - Part 1: General aspectsIEC - 479-2 Effects of current on human beings and livestock - Part 2: Special aspectsIEC - 529 Degrees of protection provided by enclosures (IP code)IEC - 644 Specification for high-voltage fuse-links for motor circuit applications


B

2.4 quality and safety of an electrical installation

Only byc the initial checking of the conformity of theelectrical installation,c the verification of the conformity of electricalequipment,c and periodic checkingcan the permanent safety of persons andsecurity of supply to equipment be achieved.

IEC - 664 Insulation coordination for equipment within low-voltage systemsIEC - 694 Common clauses for high-voltage switchgear and controlgear standardsIEC - 724 Guide to the short-circuit temperature limits of electrical cables with a rated

voltage not exceeding 0.6/1.0 kVIEC - 742 Isolation transformer and safety isolation transformer. RequirementsIEC - 755 General requirements for residual current operated protective devicesIEC - 787 Application guide for selection for fuse-links of high-voltage fuses for

transformer circuit applicationIEC - 831-1 Shunt power capacitors of the self-healing type for a.c. systems having a rated

voltage up to and including 660 V. - Part 1: General - Performance, testingand rating - Safety requirements - Guide for installation and operation


B

2. rules and statutory regulations (continued)

2.6 periodic check-testing of an installation

In many countries, all industrial andcommercial-building installations, togetherwith installations in buildings used for publicgatherings, must be re-tested periodically byauthorized agents.Table B3 shows the frequency of testingcommonly prescribed according to the kind ofinstallation concerned.

installations which require c locations at which a risk of degradation, annuallythe protection of employees fire or explosion exists

c temporary installations at worksitesc locations at which HV installations existc restrictive conducting locations where mobileequipment is usedother cases every 3 years

installations in buildings according to the type of establishmentused for public gatherings, and its capacity for receiving the public,where protection against the re-testing period will vary from onethe risks of fire and panic to three yearsare requiredresidential according to local regulations

table B3: frequency of check-tests commonly recommended for an electrical installation.

2.5 initial testing of an installation

Before a power-supply authority will connectan installation to its supply network, strictpre-commissioning electrical tests and visualinspections by the authority, or by itsappointed agent, must be satisfied.These tests are made according to local(governmental and/or institutional)regulations, which may differ slightly from onecountry to another. The principles of all suchregulations however, are common, and arebased on the observance of rigorous safetyrules in the design and realization of theinstallation.IEC 364 and related standards included inthis guide are based on an internationalconsensus for such tests, intended to coverall the safety measures and approvedinstallation practices normally required fordomestic, commercial and (the majority of)industrial buildings. Many industries howeverhave additional regulations related to aparticular product (petroleum, coal, naturalgas, etc.). Such additional requirements arebeyond the scope of this guide.The pre-commissioning electrical tests andvisual-inspection checks for installations inbuildings include, typically, all of the following:c insulation tests of all cable and wiringconductors of the fixed installation, betweenphases and between phases and earth,c continuity and conductivity tests ofprotective, equipotential and earth-bondingconductors,c resistance tests of earthing electrodes withrespect to remote earth,c allowable number of socket-outlets percircuit check,c cross-sectional-area check of allconductors for adequacy at the short-circuitlevels prevailing, taking account of theassociated protective devices, materials andinstallation conditions (in air, conduit, etc.),c verification that all exposed- andextraneous metallic parts are properlyearthed (where appropriate),c check of clearance distances in bathrooms,etc.

These tests and checks are basic (but notexhaustive) to the majority of installations,while numerous other tests and rules areincluded in the regulations to cover particularcases, for example: TN-, TT- or IT-earthedinstallations, installations based on class 2insulation, SELV circuits, and speciallocations, etc.The aim of this guide is to draw attention tothe particular features of different types ofinstallation, and to indicate the essential rulesto be observed in order to achieve asatisfactory level of quality, which will ensuresafe and trouble-free performance. Themethods recommended in this guide,modified if necessary to comply with anypossible variation imposed by a local supplyauthority, are intended to satisfy all pre-commissioning test and inspectionrequirements.


B

2.7 conformity (with standards and specifications) of equipment used in the installation

conformity of equipment with therelevant standards can be attested inseveral ways.

attestation of conformityThe conformity of equipment with the relevantstandards can be attested:c by an official conformity mark granted bythe standards organization concerned, orc by a certificate of conformity issued by alaboratory, orc by a declaration of conformity from themanufacturer.

declaration of conformityIn cases where the equipment in question isto be used by qualified or experiencedpersons, the declaration of conformityprovided by the manufacturer (included in thetechnical documentation) together with aconformity mark on the equipmentconcerned, are generally recognized as avalid attestation. Where the competence ofthe manufacturer is in doubt, a certificate ofconformity can be obtained from anindependent accredited laboratory.

mark of conformityConformity marks are inscribed onappliances and equipment which aregenerally used by technically inexperiencedpersons (for example, domestic appliances)and for whom the standards have beenestablished which permit the attribution, bythe standardization authority, of a mark ofconformity (commonly referred to as aconformity mark).

certification of QualityAssurance

A laboratory for testing samples cannot certifythe conformity of an entire production run:these tests are called type tests. In sometests for conformity to standards, the samplesare destroyed (tests on fuses, for example).Only the manufacturer can certify that thefabricated products have, in fact, thecharacteristics stated.Quality assurance certification is intended tocomplete the initial declaration or certificationof conformity.As proof that all the necessary measureshave been taken for assuring the quality ofproduction, the manufacturer obtainscertification of the quality control systemwhich monitors the fabrication of the productconcerned. These certificates are issued byorganizations specializing in quality control,and are based on the international standardISO 9000, the equivalent European standardbeing EN 29000.These standards define three model systemsof quality assurance control corresponding todifferent situations rather than to differentlevels of quality:c model 3 defines assurance of quality byinspection and checking of final products,c model 2 includes, in addition to checking ofthe final product, verification of themanufacturing process. This method applies,for example, to the manufacture of fuseswhere performance characteristics cannot bechecked without destroying the fuse,c model 1 corresponds to model 2, but withthe additional requirement that the quality ofthe design process must be rigorouslyscrutinized; for example, where it is notintended to fabricate and test a prototype(case of a custom-built product made tospecification).

the standards define severalmethods of quality assurance whichcorrespond to different situationsrather than to different levels ofquality.


B

current demandThe full-load current Ia supplied to the motoris given by the following formulae:

3-phase motor: Ia = Pn x 1,000 ex U x η x cos ϕ1-phase motor: Ia = Pn x 1,000 U x η x cos ϕwhereIa: current demand (in amps)Pn: nominal power (in kW of active power)U: voltage between phases for 3-phasemotors and voltage between the terminals forsingle-phase motors (in volts). A single-phasemotor may be connected phase-to-neutral orphase-to-phase.η: per-unit efficiency, i.e. output kW

input kWcos ϕ: power factor, i.e. kW input

kVA input

3. motor, heating and lighting loads

an examination of the actualapparent-power demands of differentloads: a necessary preliminary stepin the design of a LV installation.

The examination of actual values ofapparent-power required by each loadenables the establishment of:c a declared power demand whichdetermines the contract for the supply ofenergy,c the rating of the HV/LV transformer, whereapplicable (allowing for expected increases inload),c levels of load current at each distributionboard.

3.1 induction motors

the nominal power in kW (Pn) of amotor indicates its rated equivalentmechanical power output.The apparent power in kVA (Pa)supplied to the motor is a function ofthe output, the motor efficiency andthe power factor.

Pa = Pn η cos ϕ

motor-starting currentStarting current (Id) for 3-phase inductionmotors, according to motor type, will be:c for direct-on-line starting of squirrel-cagemotors:v Id = 4.2 to 9 In for 2-pole motorsv Id = 4.2 to 7 In for motors with more than2 poles (mean value = 6 In), whereIn = nominal full-load current of the motor,c for wound-rotor motors (with slip-rings), andfor D.C. motors:

Id depends on the value of startingresistances in the rotor circuits:Id = 1.5 to 3 In (mean value = 2.5 In).c for induction motors controlled by speed-changing variable-frequency devices (forexample: Altivar Telemecanique), assumethat the control device has the effect ofincreasing the power (kW) supplied to thecircuit motor (i.e. device plus) by 10%.

compensation of reactive-power

(kvar) supplied to induction

motors

The application of this principle to theoperation of induction motors is generallyreferred to as "power-factor improvement" or"power-factor correction".As discussed in chapter E, the apparent-power (kVA) supplied to an induction motorcan be significantly reduced by the use ofshunt-connected capacitors.Reduction of input kVA means acorresponding reduction of input current(since the voltage remains constant).Compensation of reactive-power isparticularly advised for motors that operatefor long periods at reduced power.

it is generally advantageous fortechnical and financial reasons toreduce the current supplied toinduction motors. This can beachieved by using capacitors withoutaffecting the power output of themotors.

As noted above cos ϕ = kW input so that a kVA input

reduction in kVA input will increase (i.e.improve) the value of cos ϕ.The current supplied to the motor, afterpower-factor correction, is given by:

Ia x cos ϕ cos ϕ'where cos ϕ is the power factor beforecompensation and cos ϕ' is the power factorafter compensation, Ia being the originalcurrent.

table of typical valuesTable B4 shows, as a function of the ratednominal power of motors, the currentsupplied to them at different voltage levelsunder normal uncompensated conditions,and the same motors under the sameconditions, but compensated to operate at apower factor of 0.93 (tan ϕ = 0.4).These values are averages and will differ tosome extent according to the type of motorand the manufacturer concerned.

Note: the rated voltages of certain loadslisted in table B4 are still based on 220/380 V.The international standard is now (since1983) 230/400 V.To convert the current values indicated for agiven motor rating in the 220 V and 380 Vcolumns to the currents taken by 230 V and400 V motors of the same rating, multiply bya factor of 0.95.


B

3.1 induction motors (continued)

nominal ηpowerPnkW HP %0.37 0.5 640.55 0.75 680.75 1 721.1 1.5 751.5 2 782.2 3 793 4 813.7 5 824 5.5 825.5 7.5 847.5 10 859 12 8610 13.5 8611 15 8715 20 8818.5 25 8922 30 8925 35 8930 40 8933 45 9037 50 9040 54 9145 60 9151 70 9155 75 9259 80 9263 85 9275 100 9280 110 9290 125 92100 136 92110 150 93129 175 93132 180 94140 190 94147 200 94150 205 94160 220 94180 245 94185 250 94200 270 94220 300 94250 340 94257 350 94280 380 95295 400 95300 410 95315 430 95335 450 95355 480 95375 500 95400 545 95425 580 95445 600 95450 610 95475 645 95500 680 95530 720 95560 760 95600 810 95630 855 95670 910 95710 965 95750 1020 95800 1090 95900 1220 951100 1500 95

without compensationcos ϕ Pa current at different voltagesat Pn 1-PH 3-PH

220 V 220 V 380 V 440 V 500 V 660 VkVA A A A A A A

0.73 0.79 3.6 1.8 1.03 0.99 0.91 0.60.75 1.1 4.7 2.75 1.6 1.36 1.21 0.90.75 1.4 6 3.5 2 1.68 1.5 1.10.79 1.9 8.5 4.4 2.6 2.37 2 1.50.80 2.4 12 6.1 3.5 3.06 2.6 20.80 3.5 16 8.7 5 4.42 3.8 2.80.80 4.6 21 11.5 6.6 5.77 5 3.80.80 5.6 25 13.5 7.7 7.1 5.9 4.40.80 6.1 26 14.5 8.5 7.9 6.5 4.90.83 7.9 35 20 11.5 10.4 9 6.60.83 10.6 47 27 15.5 13.7 12 8.90.85 12.3 - 32 18.5 16.9 13.9 10.60.85 13.7 - 35 20 17.9 15 11.50.86 14.7 - 39 22 20.1 18.4 140.86 19.8 - 52 30 26.5 23 17.30.86 24.2 - 64 37 32.8 28.5 21.30.86 28.7 - 75 44 39 33 25.40.86 33 - 85 52 45.3 39.4 30.30.86 39 - 103 60 51.5 45 34.60.86 43 - 113 68 58 50 390.86 48 - 126 72 64 55 420.86 51 - 134 79 67 60 440.86 57 - 150 85 76 65 490.86 65 - 170 98 83 75 570.86 70 - 182 105 90 80 610.87 74 - 195 112 97 85 660.87 79 - 203 117 109 89 690.87 94 - 240 138 125 105 820.87 100 - 260 147 131 112 860.87 112 - 295 170 146 129 980.87 125 - 325 188 162 143 1070.87 136 - 356 205 178 156 1180.87 159 - 420 242 209 184 1350.87 161 - 425 245 215 187 1400.87 171 - 450 260 227 200 1450.87 180 - 472 273 236 207 1520.87 183 - 483 280 246 210 1590.87 196 - 520 300 256 220 1700.87 220 - 578 333 289 254 1900.87 226 - 595 342 295 263 2000.88 242 - 626 370 321 281 2150.88 266 - 700 408 353 310 2350.88 302 - 800 460 401 360 2740.88 311 - 826 475 412 365 2800.88 335 - 900 510 450 400 3050.88 353 - 948 546 473 416 3200.88 359 - 980 565 481 420 3250.88 377 - 990 584 505 445 3370.88 401 - 1100 620 518 472 3650.88 425 - 1150 636 549 500 3700.88 449 - 1180 670 575 527 3950.88 478 - 1250 710 611 540 4100.88 508 - 1330 760 650 574 4450.88 532 - 1400 790 680 595 4550.88 538 - 1410 800 690 608 4600.88 568 - 1490 850 730 645 4850.88 598 - 1570 900 780 680 5150.88 634 - 1660 950 825 720 5450.88 670 - 1760 1000 870 760 5750.88 718 - 1880 1090 920 830 6300.88 754 - 1980 1100 965 850 6450.88 801 - 2100 1200 1020 910 6900.88 849 - - 1260 1075 960 7250.88 897 - - 1350 1160 1020 7700.88 957 - - 1450 1250 1100 8300.88 1076 - - 1610 1390 1220 9250.88 1316 - - 1980 1700 1500 1140

with compensationcos ϕ capa- Pa current at different voltagesat Pn citor 1-PH 3-PH

rating 220 V 220 V 380 V 440 V 500 V 660 Vkvar kVA A A A A A A

0.93 0.31 0.62 2.8 1.4 0.8 0.77 0.71 0.470.93 0.39 0.87 3.8 2.2 1.3 1.1 1 0.720.93 0.48 1.1 4.8 2.8 1.6 1.3 1.2 0.880.93 0.53 1.6 7.2 3.7 2.2 2 1.7 1.30.93 0.67 2.1 10.3 5.2 3 2.6 2.2 1.70.93 0.99 3 13.7 7.5 4.3 3.8 3.3 2.40.93 1.31 4 18 9.9 5.7 5 4.3 3.30.93 1.59 4.8 22 11.6 6.6 6.1 5.1 3.80.93 1.74 5.2 22 12.5 7.3 6.8 5.6 4.20.93 1.80 7 31 17.8 10.3 9.3 8 5.90.93 2.44 9.5 42 24 13.8 12.2 10.7 7.90.93 2.4 11.3 - 29 16.9 15.4 12.7 9.70.93 2.6 12.5 - 32 18 16.4 13.7 10.50.93 2.50 13.6 - 36 20 19 17 130.93 3.37 18.3 - 48 28 25 21 160.93 4.12 22.4 - 59 34 30 26 200.93 4.89 26.6 - 69 41 36 31 230.93 5.57 30 - 79 48 42 36 280.93 6.68 36 - 95 55 48 42 320.93 7.25 39 - 104 63 54 46 360.93 8.12 44 - 117 67 59 51 390.93 8.72 47 - 124 73 62 55 410.93 9.71 53 - 139 79 70 60 450.93 11.10 60 - 157 91 77 69 530.93 11.89 64 - 168 97 83 74 560.93 10.98 69 - 182 105 91 80 620.93 11.66 74 - 190 109 102 83 650.93 13.89 88 - 225 129 117 98 770.93 14.92 93 - 243 138 123 105 800.93 16.80 105 - 276 159 137 121 920.93 18.69 117 - 304 176 152 134 1000.93 20.24 127 - 333 192 167 146 1100.93 23.84 149 - 393 226 196 172 1260.93 24 151 - 398 229 201 175 1310.93 25.55 160 - 421 243 212 187 1360.93 26.75 168 - 442 255 221 194 1420.93 27.26 172 - 452 262 230 196 1490.93 29.15 183 - 486 281 239 206 1590.93 32.76 206 - 541 312 270 238 1780.93 33.79 212 - 557 320 276 246 1870.93 30.78 229 - 592 350 304 266 2030.93 33.81 252 - 662 386 334 293 2220.93 38.44 286 - 757 435 379 341 2590.93 39.45 294 - 782 449 390 345 2650.93 42.63 317 - 852 483 426 378 2890.93 44.80 334 - 897 517 448 394 3030.93 45.66 339 - 927 535 455 397 3060.93 47.98 356 - 937 553 478 421 3190.93 51 379 - 1041 587 490 447 3360.93 54 402 - 1088 602 519 473 3500.93 57.1 424 - 1117 634 544 499 3740.93 60.84 453 - 1183 672 578 511 3880.93 64.60 481 - 1258 719 615 543 4200.93 67.63 504 - 1325 748 643 563 4310.93 68.50 509 - 1334 757 653 575 4350.93 70.40 538 - 1410 804 691 610 4590.93 72.26 566 - 1486 852 738 643 4870.93 80.64 600 - 1571 899 781 681 5160.93 85.12 634 - 1665 946 823 719 5440.93 91.33 679 - 1779 1031 871 785 5960.93 95.81 713 - 1874 1041 913 804 6100.93 101.88 758 - 1987 1135 965 861 6530.93 107.95 804 - - 1192 1017 908 6860.93 114 849 - - 1277 1098 965 7290.93 121.68 905 - - 1372 1183 1041 7850.93 136.86 1019 - - 1523 1315 1154 8750.93 167.35 1245 - - 1874 1609 1419 1079

table B4: power and current values for typical induction motors.

Reminder: some columns refer to 220 and 380 V motors.The international (IEC 38) standard of 230/400 V has beenin force since 1983. The conversion factor for current values

for 230 V and 400 V motors is 0.95, as noted on theprevious page.


B

3. motor, heating and lighting loads (continued)

3.2. direct-current motors

D.C. motors are mainly used for specificapplications which require very high torquesand/or variable speed control (for examplemachine tools and crushers, etc.).Power to these motors is provided via speed-control converters, fed from 230/400 V3-phase a.c. sources; for example, Rectivar 4(Telemecanique).The operating principle of the converter doesnot allow heavy overloading. The speedcontroller, the supply line and the protectionare therefore based on the duty cycle of themotor (e.g. frequent starting-current peaks)rather than on the steady-state full-loadcurrent.For powers i 40 kW, this solution isprogressively replaced with a speed-changing variable-frequency device and anasynchronous motor. It is still used forgradual starters and/or retarders.

M

V power-supply network

In

Im

fig. B5: diagram of a low-power speed controller.

table B6: progressive starters with voltage ramp.

motor maximum power motor GRADIVAR catalogue number weight220 V 380 V 415 V 440 V (60 Hz) In Ith kgkW kW kW kW A A4 5.5 6 - 12 20 VR2-SA3171 3.30- - - 6.5 12 20 VR2-SA3173 3.305.5 7.5 8 - 16 30 VR2-SA3211 5.10- - - 8.5 16 30 VR2-SA3213 5.1011 18.5 20 - 37 60 VR2-SA3281 5.50- - - 21.5 37 60 VR2-SA3283 5.5018.5 30 33 - 60 100 VR2-SA3361 5.50- - - 35 60 100 VR2-SA3363 5.5022 37 40 - 72 130 VR2-SA3401 5.60- - - 42 72 130 VR2-SA3403 5.60- 55 60 - 105 200 VR2-SA3441 11.00- - - 63 105 200 VR2-SA3443 11.00- 75 80 - 140 350 VR2-SA3481 45.00- - - 90 140 350 VR2-SA3483 45.00- 132 140 - 245 530 VR2-SA3521 45.00- - - 147 245 530 VR2-SA3523 45.00

motor maximum power motor GRADIVAR catalogue number weight220 V 380 V 415 V 440 V (60 Hz) In Ith kgkW kW kW kW A A1.5 3 3.3 - 7 10 VR2-SA2121 1.95- - - 3.5 7 10 VR2-SA2123 1.954 5.5 6 - 12 20 VR2-SA2171 3.10- - - 6.5 12 20 VR2-SA2173 3.105.5 7.5 8 - 16 30 VR2-SA2211 4.90- - - 8.5 16 30 VR2-SA2213 4.9011 18.5 20 - 37 60 VR2-SA2281 5.30- - - 21.5 37 60 VR2-SA2283 5.3018.5 30 33 - 60 100 VR2-SA2361 5.30- - - 35 60 100 VR2-SA2363 5.3022 37 40 - 72 130 VR2-SA2401 5.40- - - 42 72 130 VR2-SA2403 5.40- 55 60 - 105 200 VR2-SA2441 10.00- - - 63 105 200 VR2-SA2443 10.00

table B7: progressive starters with current limitation.


B

3.3. resistive-type heating appliances and incandescent lamps (conventional or halogen)

the power consumed by a heatingappliance or an incandescent lamp isequal to the nominal power Pnquoted by the manufacturer(i.e. cos ø = 1).

the currents are given by:c 3-phase case:

Ia = Pn* ex Uc 1-phase case:

Ia = Pn* Uwhere U is the voltage between theterminals of the equipment.

The power consumed by a heating applianceor an incandescent lamp is equal to thenominal power Pn quoted by themanufacturer (i.e. cos ø = 1).The currents are given by:c 3-phase case:

Ia = Pn* ex Uc 1-phase case:

Ia = Pn* Uwhere U is the voltage between the terminalsof the equipment.

For an incandescent lamp, the use ofhalogen gas allows a more concentrated lightsource. The light output is superior and thelife of the lamp is doubled.Note: at the instant of switching on, the coldfilament gives rise to a very brief but intensepeak of current.

* Ia in amps; U in volts. Pn is in watts. If Pn isin kW, then multiply the equation by 1,000.

nominal current demandpower 1-phase 1-phase 3-phase 3-phasekW 127 V 230 V 230 V 400 V0.1 0.79 0.43 0.25 0.140.2 1.58 0.87 0.50 0.290.5 3.94 2.17 1.26 0.721 7.9 4.35 2.51 1.441.5 11.8 6.52 3.77 2.172 15.8 8.70 5.02 2.892.5 19.7 10.9 6.28 3.613 23.6 13 7.53 4.333.5 27.6 15.2 8.72 5.054 31.5 17.4 10 5.774.5 35.4 19.6 11.3 6.55 39.4 21.7 12.6 7.226 47.2 26.1 15.1 8.667 55.1 30.4 17.6 10.18 63 34.8 20.1 11.59 71 39.1 22.6 1310 79 43.5 25.1 14.4

table B8: current demands of resistive heating and incandescent lighting (conventional orhalogen) appliances.

3.4. fluorescent lamps and related equipment

the power in watts indicated on thetube of a fluorescent lamp does notinclude the power dissipated in theballast.

the current is given by:

Ia = Pballast + Pn

U x cos øIf no power-loss value is indicated forthe ballast, a figure of 25% of Pn maybe used.

standard tubular fluorescent

lamps

The power Pn (watts) indicated on the tube ofa fluorescent lamp does not include thepower dissipated in the ballast.The current taken by the complete circuit isgiven by:

Ia = Pballast + Pn U x cos øwhere U = the voltage applied to the lamp,complete with its related equipment.with (unless otherwise indicated):c cos ø = 0.6 with no power factor (PF)correction* capacitor,c cos ø = 0.86 with PF correction* (single ortwin tubes),c cos ø = 0.96 for electronic ballast.If no power-loss value is indicated for theballast, a figure of 25% of Pn may be used.Table B8 gives these values for differentarrangements of ballast.

* "Power-factor correction" is often referred to as"compensation" in discharge-lighting-tube terminology.


B

3. motor, heating and lighting loads (continued)

arrangement of tube power current (A) at 220V/240 V tubelamps, starters power consumed PF not PF electronic lengthand ballasts (W) (1) (W) corrected corrected ballast (cm)single tube with starter 18 27 0.37 0.19 60

36 45 0.43 0.24 12058 69 0.67 0.37 150

single tube without 20 33 0.41 0.21 60starter (2) with 40 54 0.45 0.26 120external starting strip 65 81 0.80 0.41 150twin tubes with starter 2 x 18 55 0.27 60

2 x 36 90 0.46 1202 x 58 138 0.72 150

twin tubes without starter 2 x 40 108 0.49 120single tube with 32 36 0.16 120high frequency ballast 50 56 0.25 150cos ø = 0.96twin tubes with high- 2 x 32 72 0.33 120frequency ballast 2 x 50 112 0.50 150cos ø = 0.96(1) Power in watts marked on tube.(2) Used exclusively during maintenance operations.

table B10: current demands and power consumption of commonly-dimensionedfluorescent lighting tubes (at 220 V/240 V - 50 Hz).

compact fluorescent tubesCompact fluorescent tubes have the samecharacteristics of economy and long life asclassical tubes.They are commonly used in public placeswhich are permanently illuminated (forexample: corridors, hallways, bars, etc.) andcan be mounted in situations otherwiseilluminated by incandescent lamps.

type of lamp lamp power current atpower consumed 220/240 V

(W) (A)globe lamps with 9 9 0.090integral ballast 13 13 0.115cos ø = 0.5 (1) 18 18 0.160

25 25 0.205electronic lamps 9 9 0.070cos ø = 0.95 (1) 11 11 0.090

15 15 0.13520 20 0.155

lamps with type 5 10 0.185starter single 7 11 0.175only "U" form 9 13 0.170incorporated cos ø ≈ 0.35 11 15 0.155(no ballast) type 10 15 0.190

double 13 18 0.165"U" form 18 23 0.220cos ø ≈ 0.45 26 31 0.315

(1) Cos ø is approximately 0.95 (the zero values of V and I are almost in phase) but the power factor is 0.5 due to theimpulsive form of the current, the peak of which occurs "late" in each half cycle.

table B11: current demands and power consumption of compact fluorescent lamps(at 220 V/240 V - 50 Hz).

3.4. fluorescent lamps and related equipment (continued)


B

3.5. discharge lamps

the power in watts indicated on thetube of a discharge lamp does notinclude the power dissipated in theballast.

These lamps depend on the luminouselectrical discharge through a gas or vapourof a metallic compound, which is contained ina hermetically-sealed transparent envelope ata pre-determined pressure.These lamps have a long start-up time,during which the current Ia is greater than thenominal current In. Power and currentdemands are given for different types of lampin table B12 (typical average values whichmay differ slightly from one manufacturer toanother).The power in watts indicated on the tube of adischarge lamp does not include the powerdissipated in the ballast.

type of power current In(A) starting luminous average utilizationlamp demand PF not PF Ia/In period efficiency life of

(W) at corrected corrected lumens lamp(W) 230V 400V 230V 400V 230V 400V (mins) (per watt) (h)high-pressure sodium vapour lamps50 60 0.76 0.3 1.4 4 to 6 80 to 120 9000 - lighting of70 80 1 0.45 to 1.6 large halls100 115 1.2 0.65 - outdoor

150 168 1.8 0.85 spaces

250 274 3 1.4 - public

400 431 4.4 2.2 lighting

1000 1055 10.45 4.9low-pressure sodium vapour lampsstandard lamp18 26.5 0.14 1.1 7 to 15 100 to 200 8000 to - lighting of35 43.5 0.62 0.24 to 1.3 12000 autoroutes

55 72 0.34 - security

90 112 0.84 0.50lighting, station

135 159 0.73platform,

180 216 0.98stockage areas

economy lamps26 34.5 0.45 0.17 1.1 7 to 15 100 to 200 8000 to - new types36 46.5 0.22 to 1.3 12000 more

66 80.5 0.39 efficient

91 105.5 0.49same

131 154 0.69utilization

mercury vapour + metal halide (also called metaliodide)70 80.5 1 0.40 1.7 3 to 5 70 to 90 6000 - lighting of150 172 1.80 0.88 6000 very large

250 276 2.10 1.35 6000 areas by

400 425 3.40 2.15 6000projectors (for

1000 1046 8.25 5.30 6000example:sports

2000 2092 2052 16.50 8.60 10.50 6 2000mercury vapour + fluorescent substance (fluorescent bulb)

stadiums, etc)

50 57 0.6 0.30 1.7 3 to 6 40 to 60 8000 to - workshops80 90 0.8 0.45 to 2 12000 with very high

125 141 1.15 0.70 ceilings (halls,

250 268 2.15 1.35hangars)

400 421 3.25 2.15- outdoor

700 731 5.4 3.85lighting

1000 1046 8.25 5.30- low light

2000 2140 2080 15 11 6.1

output (1)

(1) replaced by sodium vapour lamps.Note: these lamps are sensitive to voltage dips. They extinguish if the voltage falls to less than 50% of their nominal voltage, andwill not re-ignite before cooling for approximately 4 minutes.Note: Sodium vapour low-pressure lamps have a light-output efficiency which is superior to that of all other sources. However,use of these lamps is restricted by the fact that the yellow-orange colour emitted makes colour recognition practically impossible.

table B12: current demands of discharge lamps.

table B12 gives the current taken bya complete unit, including allassociated ancillary equipment.


B

4. power* loading of an installation

In order to design an installation, the actualmaximum load demand likely to be imposedon the power-supply system must beassessed.To base the design simply on the arithmeticsum of all the loads existing in the installationwould be extravagantly uneconomical, andbad engineering practice.The aim of this chapter is to show how allexisting and projected loads can be assignedvarious factors to account for diversity (non-simultaneous operation of all appliances of agiven group) and utilization (e.g. an electricmotor is not generally operated at its full-loadcapability, etc.). The values given are basedon experience and on records taken fromactual installations. In addition to providingbasic installation-design data on individualcircuits, the results will provide a global value

for the installation, from which therequirements of a supply system (distributionnetwork, HV/LV transformer, or generatingset) can be specified.

*power: the word "power" in the title hasbeen used in a general sense, coveringactive power (kW) apparent power (kVA) andreactive power (kvar). Where the word poweris used without further qualification in the restof the text, it means active power (kW).The magnitude of the load is adequatelyspecified by two quantities, viz:c power,c apparent power.

The ratio power = power factor apparent power

4.1 installed power (kW)

the installed power is the sum of thenominal powers of all power-consuming devices in the installation.This is not the power to be actuallysupplied in practice.

Most electrical appliances and equipmentsare marked to indicate their nominal powerrating (Pn).The installed power is the sum of the nominalpowers of all power-consuming devices in theinstallation. This is not the power to beactually supplied in practice.This is the case for electric motors, where thepower rating refers to the output power at itsdriving shaft. The input power consumptionwill evidently be greater (See 3.1).Fluorescent and discharge lamps associatedwith stabilizing ballasts, are other cases inwhich the nominal power indicated on thelamp is less than the power consumed by thelamp and its ballast (See 3.4).Methods of assessing the actual powerconsumption of motors and lightingappliances are given in Section 3 of thisChapter.The power demand (kW) is necessary tochoose the rated power of a generating set orbattery, and where the requirements of aprime mover have to be considered.For a power supply from a LV public-supplynetwork, or through a HV/LV transformer, thesignificant quantity is the apparent power inkVA.


B

From this value, the full-load currentIa (amps)* taken by the load will be:

c Ia = Pa 103 for single phase-to-neutralV connected load

c Ia = Pa 103

for three-phase balanced loadex Uwhere: V = phase-to-neutral voltage (volts)U = phase-to-phase voltage (volts)It may be noted that, strictly speaking, thetotal kVA of apparent power is not thearithmetical sum of the calculated kVA ratingsof individual loads (unless all loads are at thesame power factor).It is common practice however, to make asimple arithmetical summation, the result ofwhich will give a kVA value that exceeds thetrue value by an acceptable "design margin".

* For greater precision, account must betaken of the factor of maximum utilization asexplained below in 4-3.

4.2 installed apparent power (kVA)

The installed apparent power is commonlyassumed to be the arithmetical sum of thekVA of individual loads. The maximumestimated kVA to be supplied however is notequal to the total installed kVA.The apparent-power demand of a load (whichmight be a single appliance) is obtained fromits nominal power rating (corrected ifnecessary, as noted above for motors, etc.)and the application of the followingcoefficients:

η = the per-unit efficiency = output kWinput kW

cos ø = the power factor = kWkVA

The apparent-power kVA demand of the load

Pa = Pnη x cos ø

the installed apparent power iscommonly assumed to be thearithmetical sum of the kVA ofindividual loads. The maximumestimated kVA to be suppliedhowever is not equal to the totalinstalled kVA.

When some or all of the load characteristicsare not known, the values shown in table B13may be used to give a very approximateestimate of VA demands (individual loads are

generally too small to be expressed in kVA orkW). The estimates for lighting loads arebased on floor areas of 500 sq-metres.

fluorescent lighting (corrected to cos ø = 0.86)type of application estimated (VA/m 2) average lighting

fluorescent tube levelwith industrial reflector (1) (lux = Im/m 2)

roads and highways 7 150stockage areas, intermittent workheavy-duty works: fabrication and 14 300assembly of very large work piecesday-to-day work: 24 500office workfine work: 41 800drawing officeshigh-precision assembly workshopspower circuitstype of application estimated (VA/m 2)pumping station compressed air 3 to 6ventilation of premises 23electrical convection heaters:private houses 115 to 146flats and apartments 90offices 25dispatching workshop 50assembly workshop 70machine shop 300painting workshop 350heat-treatment plant 700

(1) example: 65 W tube (ballast not included), flux 5,100 lumens (lm), luminous efficiency of the tube = 78.5 lm/W.

table B13: estimation of installed apparent power.


B

4. power* loading of an installation (continued)

4.3 estimation of actual maximum kVA demand

all individual loads are notnecessarily operating at full ratednominal power nor necessarily at thesame time. Factors ku and ks allowthe determination of the maximumpower and apparent-power demandsactually required to dimension theinstallation.

All individual loads are not necessarilyoperating at full rated nominal power nornecessarily at the same time. Factors ku andks allow the determination of the maximumpower and apparent-power demands actuallyrequired to dimension the installation.

factor of maximum utilization (ku)In normal operating conditions the powerconsumption of a load is sometimes less thanthat indicated as its nominal power rating, afairly common occurrence that justifies theapplication of an utilization factor (ku) in theestimation of realistic values.This factor must be applied to each individualload, with particular attention to electricmotors, which are very rarely operated at fullload.

In an industrial installation this factor may beestimated on an average at 0.75 for motors.For incandescent-lighting loads, the factoralways equals 1.For socket-outlet circuits, the factors dependentirely on the type of appliances beingsupplied from the sockets concerned.

factor of simultaneity (ks)It is a matter of common experience that thesimultaneous operation of all installed loadsof a given installation never occurs inpractice, i.e. there is always some degree ofdiversity and this fact is taken into account forestimating purposes by the use of asimultaneity factor (ks).The factor ks is applied to each group ofloads (e.g. being supplied from a distributionor sub-distribution board).The determination of these factors is theresponsibility of the designer, since it requiresa detailed knowledge of the installation andthe conditions in which the individual circuitsare to be exploited. For this reason, it is notpossible to give precise values for generalapplication.

Factor of simultaneity for an apartmentblockSome typical values for this case are given intable B14, and are applicable to domesticconsumers supplied at 230/400 V (3-phase4-wires). In the case of consumers usingelectrical heat-storage units for spaceheating, a factor of 0.8 is recommended,regardless of the number of consumers.Example:5 storeys apartment building with25 consumers, each having 6 kVA of installedload.The total installed load for the building= 36 + 24 + 30 + 36 + 24 = 150 kVAThe apparent-power supply required for thebuilding = 150 x 0.46 = 69 kVAFrom table B 14, it is possible to determinethe magnitude of currents in different sectionsof the common main feeder supplying allfloors. For vertical rising mains fed at groundlevel, the cross-sectional area of theconductors can evidently be progressivelyreduced from the lower floors towards theupper floors.These changes of conductor size areconventionally spaced by at least 3-floorintervals.In the example, the current entering the risingmain at ground level is150 x 0.46 x 103

400 x e = 100 A

The current entering the third floor is:(36+24) x 0.63 x 103

400 x e = 55 A

number of down- factor ofstream consumers simultaneity (ks)2 to 4 15 to 9 0.7810 to 14 0.6315 to 19 0.5320 to 24 0.4925 to 29 0.4630 to 34 0.4435 to 39 0.4240 to 49 0.4150 and more 0.40

table B14: simultaneity factors in anapartment block.

4thfloor

6 consumers36 kVA

3rdfloor

2ndfloor

1stfloor

groundfloor

4 consumers24 kVA

6 consumers36 kVA

5 consumers30 kVA

4 consumers24 kVA

0.78

0.63

0.53

0.49

0.46

fig. B15: application of the factor ofsimultaneity (ks) to an apartment block of5 storeys.


B

n°2

1

distributionbox

workshop A 5 0.8

0.8

0,8

0.8

0.8

0.8

5

5

5

2

2

lathe

18

3 11

10.8

0.41

15

10.6

2,5

2,5

15

15

ventilationfan

0.28118

112

1oven

n°2

n°4n°3

30 fluorescentlamps

pedestal-drill

workshop B compressor

workshop C

n°2

n°2n°1

n°1

n°1

4

4

4

4

1.6

1.6

18

3

14.4

12

1

1

1

1

2.5

2

18

15

15

2.5

workshop Adistributionboard

0.75powercircuit

workshopBdistributionboard

workshopCdistributionboard

maingeneraldistributionboardMGDB

socket-outlets

lightingcircuit

0,9

0.9

0.9

0.910.6

3,6

3

12

4,31

35

5

2

65LV/HV

distributionbox

111

0.21

n°1

10/16 A5 socket-outlets

powercircuit

socket-outletslightingcircuit

socket-outlets

lightingcircuit

powercircuit

10 fluorescentlamps

3 socket-outlets 10/16 A

20 fluorescentlamps

10/16 A5 socket-outlets

37.8

15.6

18.9

Factor of simultaneity for distributionboardsTable B16 shows hypothetical values of ks fora distribution board supplying a number ofcircuits for which there is no indication of themanner in which the total load dividesbetween them.If the circuits are mainly for lighting loads, it isprudent to adopt ks values close to unity.

number of factor ofcircuits simultaneity (ks)assemblies entirely tested2 and 3 0.94 and 5 0.86 to 9 0.710 and more 0.6assemblies partially testedin every case choose 1.0

table B16: factor of simultaneity fordistribution boards (IEC 439).

Factor of simultaneity accordingto circuit functionks factors which may be used for circuitssupplying commonly-occurring loads, areshown in table B17.

circuit function factor ofsimultaneity (ks)

lighting 1heating and airconditioning 1socket-outlets 0.1 to 0.2 (1)lifts and catering hoists (2)- for the most powerfulmotor 1- for the second mostpowerful motor 0.75- for all other motors 0.60

(1) In certain cases, notably in industrial installations, thisfactor can be higher.(2) The current to take into consideration is equal to thenominal current of the motor, increased by a third of itsstarting current.

table B17: factor of simultaneity accordingto circuit function.

4.4 example of application of factors ku and ks

an example in the estimation ofactual maximum kVA demands at alllevels of an installation, from eachload position to the point of supply.

In this example, the total installed apparentpower is 126.6 kVA, which corresponds to anactual (estimated) maximum value at the LVterminals of the HV/LV transformer of 65 kVAonly.Note: in order to select cable sizes for thedistribution circuits of an installation, thecurrent I (in amps) through a circuit isdetermined from the equation

I = kVA x 103

U ewhere kVA is the actual maximum 3-phaseapparent-power value shown on the diagramfor the circuit concerned, and U is the phase-to-phase voltage (in volts).

level 1 level 2 level 3

utilization apparent- utilization apparent- simultaneity apparent- simultaneity apparent- simultaneity apparent-power (Pa) factor power demand factor power demand factor power demand factor power demandkVA max. max. kVA kVA kVA kVA

table B18: an example in estimating the maximum predicted loading of an installation (the factor values used are for demonstration purposesonly).


B

In (A)voltage (at no load) 400 V 420 V 433 V 480 Vrated power (kVA)50 72 69 67 60100 144 137 133 120160 231 220 213 192250 361 344 333 301315 455 433 420 379400 577 550 533 481500 722 687 667 601630 909 866 840 758800 1155 1100 1067 9621000 1443 1375 1333 12031250 1804 1718 1667 15041600 2309 2199 2133 19252000 2887 2749 2667 24062500 3608 3437 3333 3007

4. power loading of an installation (continued)

4.5 diversity factor

The term DIVERSITY FACTOR, as definedin IEC standards, is identical to the factor ofsimultaneity (ks) used in this guide, asdescribed in 4.3. In some English-speakingcountries however (at the time of writing)DIVERSITY FACTOR is the inverse of ks i.e.it is always u 1.

4.6 choice of transformer rating

When an installation is to be supplied directlyfrom a HV/LV transformer and the maximumapparent-power loading of the installation hasbeen determined, a suitable rating for thetransformer can be decided, taking dueaccount of the following considerations:

c the possibility of improving the power factorof the installation (see chapter E),c anticipated extensions to the installation,c installation constraints (temperature...)standard transformer ratings.

table B19: IEC-standardized kVA ratings of HV/LV 3-phase distribution transformers andcorresponding nominal full-load current values.

The nominal full-load current In on the LVside of a 3-phase transformer is given by:

In = Pa 103 where UePa = kVA rating of the transformerU = phase-to-phase voltage at no-load* (involts)In is in amperes.For a single-phase transformer:

In = Pa 103 where VV = voltage between LV terminals at no-load*(in volts).

Simplified equation for 400 V (3-phase load)In = kVA x 1.4The IEC standard for power transformers isIEC 76.* as given on the transformer-ratingnameplate. For table B19 the no-load voltageused is 420 V for the nominal 400 V winding.


B

4.7 choice of power-supply sources

The study developed in F2 on the importanceof maintaining a continuous supply raises thequestion of the use of standby-power plant.The choice and characteristics of thesealternative sources are described in F3-3.For the main source of supply the choice isgenerally between a connection to the HV orthe LV network of the public power-supplyauthority.In practice, connection to a HV source maybe necessary where the load exceeds (oris planned eventually to exceed) a certainlevel - generally of the order of 250 kVA, or ifthe quality of service required is greater thanthat normally available from a LV network.Moreover, if the installation is likely to causedisturbance to neighbouring consumers,when connected to a LV network, the supplyauthorities may propose a HV service.Supplies at HV can have certain advantages:in fact, a HV consumer:c is not disturbed by other consumers, whichcould be the case at LV,c is free to choose any type of LV earthingsystem,c has a wider choice of economic tariffs,c can accept very large increases in load.

It should be noted, however, that:c the consumer is the proprietor of the HV/LVsubstation and, in some countries, he mustbuild and equip it at his own expense. Thepower authority can, in certain circumstances,participate in the investment, at the level ofthe HV line for example,c a part of the connection costs can, forinstance, often be recovered if a secondconsumer is connected to the HV line within acertain time following the original consumer'sown connection,c the consumer has access only to the LVpart of the installation, access to the HV partbeing reserved to the supply-authoritypersonnel (meter reading, operationalmanœuvres, etc.). However, in certaincountries, the HV protective circuit breaker(or fused load-break switch) can be operatedby the consumer,c the type and location of the substation areagreed between the consumer and thesupply authority.

HV/LV distribution substations

C

HV/LV distribution substations

1. Supply of power at high voltage1.1 power supply characteristics of high-voltage distribution networks C11.2 different HV service connections C111.3 some operational aspects of HV distribution networks C13

2. Consumers HV substations2.1 procedures for the establishment of a new substation C15

3. Substation protection schemes3.1 protection against electric shocks and overvoltages C173.2 electrical protection C223.3 protection against thermal effects C313.4 interlocks and conditioned manœuvres C31

4. The consumer substation with LV metering4.1 general C344.2 choice of panels C364.3 choice of HV switchgear panel for a transformer circuit C384.4 choice of HV/LV transformer C38

5. A consumer substation with HV metering5.1 general C445.2 choice of panels C465.3 parallel operation of transformers C48

6. Constitution of HV/LV distribution substations6.1 different types of substation C496.2 indoor substations equipped with metal-enclosed switchgear C496.3 outdoor substations C52

HV/LV distribution substations - C1

C

1. supply of power at high voltage

At present there is no international agreementon precise limits to define “High” voltage.Voltage levels which are designated as “high”in some countries are referred to as“medium” in others.In this chapter, distribution networks whichoperate at voltages of 1,000 V or less arereferred to as Low-Voltage systems, whilesystems of power distribution which require

one stage of stepdown voltagetransformation, in order to feed into low-voltage networks, will be referred to as High-Voltage systems.For economic and technical reasons theupper nominal voltage limit of high-voltagedistribution systems, as defined above,seldom exceeds 36.5 kV.

1.1 power-supply characteristics of high voltage distribution networks

the main features which characterizea power-supply system include:c the nominal voltage and relatedinsulation levels,c the short-circuit current,c the rated normal current of items ofplant and equipment,c the method of earthing.Note: All voltages and currents arer.m.s. values, unless otherwisestated.

in this document, the word “nominal”voltage is used for the network andthe word “rated” voltage is used forthe equipment.

nominal voltage and related

insulation levels

The nominal voltage of a system or of anequipment is defined in IEC 38 as “thevoltage by which a system or equipment isdesignated and to which certain operatingcharacteristics are referred”. Closely relatedto the nominal voltage is the “highest voltagefor equipment” which concerns the level ofinsulation at normal working frequency, andto which other characteristics may be referredin relevant equipment recommendations.The “highest voltage for equipment” isdefined in IEC 38 as:“the maximum value of voltage for which theequipment may be used, that occurs undernormal operating conditions at any time andat any point on the system. It excludesvoltage transients, such as those due tosystem switching, and temporary voltagevariations”.

Notes:1.- The highest voltage for equipment isindicated for nominal system voltages higherthan 1,000 V only. It is understood that,particularly for certain nominal systemvoltages, normal operation of equipmentcannot be ensured up to this highest voltagefor equipment, having regard to voltagesensitive characteristics such as losses ofcapacitors, magnetizing current oftransformers, etc.In such cases, the relevant recommendationsmust specify the limit to which the normaloperation of this equipment can be ensured.2.- It is understood that the equipment to beused in systems having nominal voltage notexceeding 1,000 V should be specified withreference to the nominal system voltage only,both for operation and for insulation.3.- The definition for “highest voltage forequipment” given in IEC 38 is identical to thatgiven in IEC 694 for “rated voltage”.IEC 694 concerns switchgear for nominalvoltages exceeding 1,000 V.

C2 - HV/LV distribution substations

C

1. supply of power at high voltage (continued)

1.1 power-supply characteristics of high voltage distribution networks (continued)

The following Table C1, taken from IEC 38,lists the most-commonly used standard levelsof high-voltage distribution, and relates thenominal voltages to corresponding standardvalues of “Highest Voltage for Equipment”.Two series of highest voltages for equipmentare given below, one for 50 Hz and 60 Hzsystems (Series I), the other for 60 Hzsystems (Series II - North American practice).It is recommended that only one of theseseries should be used in any one country.It is also recommended that only one of thetwo series of nominal voltages given forSeries I should be used in any one country.These systems are generally three-wiresystems unless otherwise indicated. Thevalues shown are voltages between phases.The values indicated in parentheses shouldbe considered as non-preferred values. It isrecommended that these values should notbe used for new systems to be constructed infuture.

Notes:1 - It is recommended that in any one countrythe ratio between two adjacent nominalvoltages should be not less than two.2 - In a normal system of Series I, the highestvoltage and the lowest voltage do not differby more than approximately ± 10% from thenominal voltage of the system. In a normalsystem of Series II, the highest voltage doesnot differ by more than + 5% and the lowestvoltage by more than - 10% from the nominalvoltage of the system.

series I series IIhighest voltage nominal system highest voltage nominal systemfor equipement voltage for equipment voltage(kV) (kV) (kV) (kV)3.6 (1) 3.3 (1) 3 (1) 4.40 (1) 4.16 (1)

7.2 (1) 6.6 (1) 6 (1) - -12 11 10 - -- - - 13.2 (2) 12.47 (2)

- - - 13.97 (2) 13.2 (2)

- - - 14.52 (1) 13.8 (1)

(17.5) - (15) - -24 22 20 - -- - - 26.4 (2) 24.94 (2)

36 (3) 33 (3) - - -- - - 36.5 (2) 34.5 (2)

40.5 (3) - 35 (3) - -

1) These values should not be used for public distribution systems.2) These systems are generally four-wire systems.3) The unification of these values is under consideration.

table C1: relating nominal system voltages with corresponding rated system voltages(r.m.s. values).

In order to ensure adequate protection ofequipment against abnormally-high short-term power-frequency overvoltages, andtransient overvoltages caused by lightning,switching, and system fault conditions, etc.all HV equipment must be specified to haveappropriate Rated Insulation Levels.

SwitchgearTable C2 shown below, is extracted fromIEC 694 and lists standard values of“withstand” voltage requirements. The choicebetween List 1 and List 2 values of table C2depends on the degree of exposure tolightning and switching overvoltages*, thetype of neutral earthing, and the type ofovervoltage protection devices, etc. (forfurther guidance reference should be made toIEC 71).* This means basically that List 1 generally applies toswitchgear to be used on underground-cable systems whileList 2 is chosen for switchgear to be used on overhead-linesystems.


CBased on current practice in most Europeanand several other countries

rated rated lightning impulse withstand voltage rated I min power-voltage (peak value) frequency withstandU list 1 list 2 voltage (r.m.s. value)(r.m.s. to earth, across the to earth, across the to earth, across thevalue) between poles isolating between poles isolating between poles isolating

and across distance and across distance and across distanceopen open openswitching switching switchingdevice device device

(kV) (kV) (kV) (kV) (kV) (kV) (kV)3.6 20 23 40 46 10 127.2 40 46 60 70 20 2312 60 70 75 85 28 3217.5 75 85 95 110 38 4524 95 110 125 145 50 6036 145 165 170 195 70 8052 - - 250 290 95 11072.5 - - 325 375 140 160

Note: The withstand voltage values “across the isolating distance” are valid only for the switchingdevices where the clearance between open contacts is designed to meet safety requirementsspecified for disconnectors (isolators).

table C2: switchgear rated insulation levels.

It should be noted that, at the voltage levelsin question, no switching overvoltage ratingsare mentioned. This is because overvoltagesdue to switching transients are less severe atthese voltage levels than those due tolightning.

TransformersThe two tables C3A and C3B shown belowhave been extracted from IEC 76-3, and referto the current practices in countries otherthan those of North America (Series I) and tothose of North America and some othercountries (Series II).The significance of list 1 and list 2 in Series Iis the same as that for the switchgear table,i.e. the choice depends on the degree ofexposure to lightning, etc.

highest voltage rated short duration rated lightning impulsefor equipment Um power frequency withstand voltage(r.m.s.) withstand voltage (peak)

(r.m.s.) list 1 list 2(kV) (kV) (kV) (kV)i 1.1 3 - -3.6 10 20 407.2 20 40 6012 28 60 7517.5 38 75 9524 50 95 12536 70 145 17052 95 25072.5 140 325

table C3A: transformers rated insulation levels in series I (based on current practice otherthan in the United States of America and some other countries).


C



highest voltage rated short duration rated lightning impulsefor equipment Um power frequency withstand voltage(r.m.s.) withstand voltage (peak)

(r.m.s.) distribution othertransformers transformers

(kV) (kV) (kV) (kV)4.40 19 60 7513.2013.97 34 95 11014.5226.4 50 15036.5 70 20072.5 140 350

table C3B: transformers rated insulation levels in series II (based on current practice inthe United States of America and some other countries).

Other componentsIt is evident that the insulation performance ofother HV components associated with thesemajor items, e.g. porcelain or glassinsulators, HV cables, instrumenttransformers, etc. must be compatible withthat of the switchgear and transformers notedabove. Test schedules for these items aregiven in appropriate IEC publications.

General noteThe IEC standards are intended for world-wide application and consequently embracean extensive range of voltage and currentlevels.These reflect the diverse practices adopted incountries of different meteorologic,geographic and economic constraints.The national standards of any particularcountry are normally rationalized to includeone or two levels only of voltage, current, andfault-levels, etc.

the national standards of anyparticular country are normallyrationalized to include one or twolevels only of voltage, current, andfault-levels, etc.

short-circuit current

A circuit breaker (or fuse switch, over alimited voltage range) is the only form ofswitchgear capable of safely breaking thevery high levels of current associated withshort-circuit faults occurring on a powersystem.Standard values of circuit breaker short-circuit current-breaking capability arenormally given in kilo-amps.These values refer to a 3-phase short-circuitcondition, and are expressed as the averageof the r.m.s. values of the a.c. component ofcurrent in each of the three phases.

Short-circuit current-breaking ratingsFor circuit breakers in the rated voltageranges being considered in this chapter,IEC 56 gives standard short-circuit current-breaking ratings as follows.

a circuit breaker (or fuse switch, overa limited voltage range) is the onlyform of switchgear capable of safelybreaking the very high levels ofcurrent associated with short-circuitfaults occurring on a power system.

kV 3.6 7.2 12 17.5 24 36 52kA 8 8 8 8 8 8(r.m.s.) 10 12.5 12.5 12.5 12.5 12.5 12.5

16 16 16 16 16 16 2025 25 25 25 25 2540 40 40 40 40 40

50

table C4: standard short-circuit current-breaking ratings extracted from table X IEC 56.


CWhere the installation of a circuit breaker iselectrically remote from a power source, it isonly necessary to check that the power factorof the faulty circuit is not less than 0.07 andthat the minimum operating time of protectiverelaying is not less than a half cycle of thepower-supply frequency (i.e. 10 ms at 50 Hz).In the great majority of cases, theseconditions will be satisfied in a conventionalHV distribution network.In such circumstances, it is then onlynecessary to ensure that the IEC-rated short-circuit current-breaking capability of the circuitbreaker exceeds the r.m.s. value of 3-phaseshort-circuit current at the point of installation.Where circuit breakers are to be installedclose to generating plant, the a.c. componentof short-circuit current will diminish rapidlyfrom its initial value (i.e. the a.c. decrement)and the power factor of the fault circuit maybe less than 0.07. Such a case would needfurther investigation along the lines indicatedin IEC 56, since the result could lead to anabsence of current zeros for several initialcycles*.

Maximum peak of currentAnother aspect of short-circuit currentstresses that may be imposed on thecomponent parts of a power system,concerns the maximum possible peak ofcurrent which can occur if a circuit breaker isclosed on to a dead circuit which is short-circuited.For such a possibility, circuit breakers have ashort-circuit current-making rating, expressedin kA of peak current. The numerical value ofthis rating is 2.5 times the short-circuitcurrent-breaking rating of the circuit breaker.ExplanationThe value 2.5 is derived as follows: short-circuit current is normally highly inductive sothat at least two of the phases will contain atransient d.c. component. In the worstpossible case, the value of the d.c.component in one of the phases will be equal

to the peak value of the a.c. component,producing the so-called “doubling effect”.However, the d.c. transient diminishes rapidlyfrom the instant of fault, while the peakcurrent occurs a half cycle after that instant.Allowance is made for the diminution in thed.c. component by reducing the doublingfactor (2) to a value of 1.8.In IEC 56 this reduction is based on aninductive d.c. time constant value which isrepresentative of average HV distributionsystems.The peak current value is thereforerIrms x 1.8 = 2.54 Irms which is rounded offfor standardization purposes to 2.5 Irms.The form of the fully-offset short-circuitcurrent is shown in figure C5, reproducedfrom IEC 56.

Note:When a short-circuit (s.c.) occurs on a powersystem, all electric motors act for a very briefperiod (1-2 cycles) as generators, and feedcurrent (typically 50 % - 80 % of the motor-starting current) into the fault. This is due tothe collapsing magnetic flux in each motorand is generally significant only for the firstpower-frequency cycle from the moment ofs.c.For the latter reason, apart from veryexceptional cases, it is not necessary to takeaccount of its effect on the s.c. current-breaking rating of a circuit breaker (CB).It cannot be neglected however, in the caseof s.c. current-making rating.If there are large concentrations of motorsnear the point of installation of a CB, the s.c.current-making level will be greater than2.5 times the s.c. current-breaking level at thesame location.In order to ensure an adequate s.c. current-making capacity, therefore, a CB having anoversized s.c. current-breaking capacity isnecessary in such circumstances.* A "natural" current zero is essential for the correctfunctioning of a CB, unless especially designed for thepurpose.

B

IAC

IDC

IMC

A

D

C

B’

A’

D’

C’

t

X

E’

E

I

IMC = making current = (A-C) 1.8 where Aand C are measured at t = 0

IAC = peak value of a.c. component ofcurrent at instant EE’

IAC = r.m.s. value of the a.c. component r of current at instant EE’IDC = d.c. component of current at instant

EE’IDC x 100 IAC

= percentage value of the d.c. component

AA'BB’ = envelope of current-wave

BX = normal zero lineCC’ = displacement of current-wave zero-

line at any instantDD’ = r.m.s. value of the a.c. component of

current at any instant, measured fromCC’

EE’ = instant of contact separation (initiationof the arc)

fig. C5: determination of short-circuit making and breaking currents, and of percentaged.c. component.


C



the short-circuit current level of a HVdistribution system is frequentlylimited by design techniques to a pre-determined maximum value typicallyin the range of 12.5 kA to 25 kA.All HV equipments connected to thesystem must be capable ofwithstanding, without damage, thethermal and mechanical stresses ofthe maximum short-circuit current for1 second, or in particular cases(depending on equipmentspecifications) for 3 seconds.

In such a case, a CB having a s.c. current-breaking rating sufficiently high to ensure anadequate s.c. current-making performancemust be installed.The short-circuit current level of a HVdistribution system is frequently limited bydesign techniques to a pre-determinedmaximum value typically in the range of12.5 kA to 25 kA.All HV equipments connected to the systemmust be capable of withstanding, withoutdamage, the thermal and mechanicalstresses of the maximum short-circuit currentfor 1 second, or in particular cases(depending on equipment specifications) for3 seconds.

the most common normal currentrating for general-purpose HVdistribution switchgear is 400 A.

Rated normal currentThe rated normal current is defined as “ther.m.s. value of the current which can becarried continuously at rated frequency with atemperature rise not exceeding that specifiedby the relevant product standard”.The rated normal current requirements forswitchgear are decided at the substationdesign stage.The most common normal current rating forgeneral-purpose HV distribution switchgear is400 A.In industrial areas and high-load-densityurban districts, circuits rated at 630 A aresometimes required, while at bulk-supplysubstations which feed into HV networks,800 A; 1,250 A; 1,600 A; 2,500 A and 4,000 Acircuit breakers are listed in IEC 56 asstandard ratings for incoming-transformercircuits, bus-section and bus-coupler CBs,etc.At HV/LV substations which include one (ormore) transformer(s) with a normal primarycurrent of less than 45 A, a HV switchassociated with a set of 3 fuses (or acombination switch-fuse) is generally used tocontrol and protect the transformer, as a moreeconomic alternative to a CB.There are no IEC-recommended normal-current rating tables for the combination inthese cases. The actual rating will be givenby the switch-fuse manufacturer, according tothe fuse characteristics, and details of thetransformer, such as:c normal current at HV,c permissible overcurrent and its duration,c max. peak and duration of the transformerenergization inrush magnetizing current,c off-circuit tapping-switch position, etc. asshown in the example given in Appendix A ofIEC 420, and summarized in Appendix C1 ofthis guide.

The IEC recommends that the normal-currentrating value, assigned to the combination bythe manufacturer, be one of the “R10” seriesof (ISO) preferred numbers, viz: 10, 12.5, 16,20, 25, 31.5, 40, 50, 63, 80 with multiples (orsub-multiples) of 10 as required.In such a scheme, the load-break switchmust be suitably rated to trip automatically,e.g. by relays, at low fault-current levelswhich must cover (by an appropriate margin)the rated minimum breaking current of the HVfuses. In this way, high values of fault currentwhich are beyond the breaking capability ofthe load-break switch will be cleared by thefuses, while low fault-current values, thatcannot be correctly broken by the fuses, willbe cleared by the relay-operated load-breakswitch.Appendix C1 gives further information on thisarrangement, as applied to HV switch-fusecombination units.


CInfluence of the ambient temperature andaltitude on the rated currentNormal-current ratings are assigned to allcurrent-carrying electrical appliances, andupper limits are decided by the acceptabletemperature rise caused by the I2R (watts)dissipated in the conductors, (where I = r.m.s.current in amperes and R = the resistance ofthe conductor in ohms), together with theheat produced by magnetic-hysteresis andeddy-current losses in motors, transformers,etc. and dielectric losses in cables andcapacitors, where appropriate.The temperature rise above the ambienttemperature will depend mainly on the rate atwhich the heat is removed. For example,large currents can be passed through electricmotor windings without causing them tooverheat, simply because a cooling fan fixedto the shaft of the motor removes the heat atthe same rate as it is produced, and so thetemperature reaches a stable value belowthat which could damage the insulation and

result in a burnt-out motor.Oil- and/or air-cooled transformers are amongthe most widely known examples of such“forced-cooling” techniques.The normal-current values recommended byIEC are based on ambient-air temperaturescommon to temperate climates at altitudesnot exceeding 1,000 metres, so that itemswhich depend on natural cooling by radiationand air-convection will overheat if operated atrated normal current in a tropical climate and/or at altitudes exceeding 1,000 metres. Insuch cases, the equipment has to be derated,i.e. be assigned a lower value of normal-current rating according to IEC 76-2.In the case of force-cooled transformers it isgenerally sufficient to provide sun shields,and increase the oil-cooling radiator surfaces,the amount of cooling oil, the power of thecirculating-oil pumps, and the size of the air-circulating fans, to maintain the original IECrating.

earth faults on high-voltage systemscan produce dangerous voltagelevels on LV installations. LVconsumers (and substation operatingpersonnel) can be safeguardedagainst this danger by:c restricting the magnitude of HVearth-fault currents,c reducing the substation earthingresistance to the lowest possiblevalue,c creating equipotential conditions atthe substation and at the consumer'sinstallation.

earthing connections

Earth faults on high-voltage systems canproduce dangerous voltage levels on LVinstallations. LV consumers (and substationoperating personnel) can be safeguardedagainst this danger by:c restricting the magnitude of HV earth-faultcurrents,c reducing the substation earthing resistanceto the lowest possible value,c creating equipotential conditions at thesubstation and at the consumer's installation.Earthing and equipment-bonding earthconnections require careful consideration,particularly regarding safety of the LVconsumer during the occurrence of a short-circuit to earth on the HV system.

Earth electrodesIn general, it is preferable, where physicallypossible, to separate the electrode providedfor earthing exposed conductive parts of HVequipment from the electrode intended forearthing the LV neutral conductor. This iscommonly practised in rural systems wherethe LV neutral-conductor earth electrode isinstalled at one or two spans of LVdistribution line away from the substation.In most cases, the limited space available inurban substations precludes this practice, i.e.there is no possibility of separating a HVelectrode sufficiently from a LV electrode toavoid the transference of (possiblydangerous) voltages to the LV system.

Earth-fault currentEarth-fault current levels at high voltage aregenerally (unless deliberately restricted)comparable to those of a 3-phase short-circuit.Such currents passing through an earthelectrode will raise its voltage to a high valuewith respect to “remote earth” (the earthsurrounding the electrode will be raised to ahigh potential; “remote earth” is at zeropotential).For example, 10,000 A of earth-fault currentpassing through an electrode with an(unusually low) resistance of 0.5 ohms willraise its voltage to 5,000 V.Providing that all exposed metal in thesubstation is “bonded” (connected together)and then connected to the earth electrode,

and the electrode is in the form of (or isconnected to) a grid of conductors under thefloor of the substation, then there is nodanger to personnel, since this arrangementforms an equipotential “cage” in which allconductive material, including personnel, israised to the same potential.

Transferred potentialA danger exists however from the problemknown as Transferred Potential. It will beseen in figure C6 that the neutral point of theLV winding of the HV/LV transformer is alsoconnected to the common substation earthelectrode, so that the neutral conductor, theLV phase windings and all phase conductorsare also raised to the electrode potential.Low-voltage distribution cables leaving thesubstation will transfer this potential toconsumers installations. It may be noted thatthere will be no LV insulation failure betweenphases or from phase to neutral since theyare all at the same potential. It is probable,however, that the insulation between phaseand earth of a cable or some part of aninstallation would fail.

SolutionsA first step in minimizing the obvious dangersof transferred potentials is to reduce themagnitude of HV earth-fault currents. This iscommonly achieved by earthing the HVsystem through resistors or reactors at thestar points of selected transformers*, locatedat bulk-supply substations.A relatively high transferred potential cannotbe entirely avoided by this means, however,and so the following strategy has beenadopted in some countries.The equipotential earthing installation at aconsumer's premises represents a remoteearth, i.e. at zero potential. However, if thisearthing installation were to be connected bya low-impedance conductor to the earth-electrode at the substation, then theequipotential conditions existing in thesubstation would also exist at the consumer'sinstallation.* the others being unearthed. A particular case of earth-faultcurrent limitation, namely, by means of a Petersen coil, isdiscussed at the end of Sub-clause 3.2.


C



N

3

2

1

fault

HV LV

If

If

Rs

V = IfRs

consumer

fig. C6: transferred potential.

Low-impedance interconnectionThis low-impedance interconnection isachieved simply by connecting the neutralconductor to the consumer's equipotentialinstallation, and the result is recognized asthe TN system of earthing (IEC 364-3) asshown in diagram A of figure C7.The TN system is generally associated with aProtective Multiple Earthing (PME) scheme,in which the neutral conductor is earthed atintervals along its length (every 3rd or 4thpole on a LV overhead-line distributor) and ateach consumer's service position.It can be seen that the network of neutralconductors radiating from a substation, eachof which is earthed at regular intervals,constitutes, together with the substationearthing, a very effective low-resistance earthelectrode.The combination of restricted earth-faultcurrents, equipotential installations and low-resistance substation earthing, results ingreatly reduced levels of overvoltage andlimited stressing of phase-to-earth insulationduring the type of HV earth-fault situationdescribed above.

Limitation of the HV earth-fault currentand earth resistance of the substationAnother widely-used system of earthing isshown in diagram C of figure C7. It will beseen that in the TT system, the consumer'searthing installation (being isolated from thatof the substation) constitutes a remote earth.This means that, although the transferredpotential will not stress the phase-to-phaseinsulation of the consumer's equipment, thephase-to-earth insulation of all three phaseswill be subjected to overvoltage.The strategy in this case, is to:c restrict the value of HV earth-fault currents,as previously discussed,c reduce the resistance of the substationearth electrode, such that the standard valueof 5-second withstand-voltage-to-earth for LVequipment and appliances will not beexceeded.


C

N

3

2

1

HV LV

TN(R)

N

3

2

1

HV LV

RS

IT(R)

cases A and B

No particular resistance value for Rs isimposed in these cases.

A B

cases C and D

WhereUw = the rated normal-frequency withstand

voltage for low-voltage equipment atconsumer installations

Uo = phase to neutral voltage at consumer'sinstallations

Im = maximum value of HV earth-fault current

Rs i Uw - Uo Im

N

3

2

1

HV LV

RS

TT(N)

N

3

2

1

HV LV

RS

IT(N)

C D

N

3

2

1

HV LV

RNRS

TT(S)

N

3

2

1

HV LV

RN

IT(S)

RSE F

In cases E and F the LV protective conductors (bonding exposed conductive parts)in the substation are earthed via the substation earth electrode, and it is therefore the substationLV equipment (only) that could be subjected to overvoltage.

cases E and F

WhereUws = the normal-frequency withstand

voltage for low-voltage equipments inthe substation (since the exposedconductive parts of these equipmentsare earthed via Rs)

U = phase to neutral voltage at the substationfor the TT(s) system, but the phase-to-phase voltage for the IT(s) system

Im = maximum value of HV earth-fault current

Rs i Uws - U Im

Notes:(R) signifies that the HV and LV exposed conductive parts at the substation and those at the consumer's installations, together with the LV neutralpoint of the transformer, are all earthed via the substation electrode system.(N) signifies that the HV and LV exposed conductive parts at the substation, together with the LV neutral point of the transformer are earthed via thesubstation electrode system.(S) signifies that the LV neutral point of the transformer is separately earthed outside of the area of influence of the substation earth electrode.Uw and Uws are commonly given the (IEC 644) value 1.5 Uo + 750 V, where Uo is the nominal phase-to-neutral voltage of the LV system concerned.

fig. C7: maximum earthing resistance Rs at a HV/LV substation to ensure safety during a short-circuit to earth fault on the hig h-voltageequipment for different systems of earthing.


C



Practical values adopted by one nationalelectrical power-supply authority, on its 20 kVdistribution systems, are as follows:c maximum earth-fault current on overhead-line distribution systems, or mixed (O/H lineand U/G cable) systems, is 300 A,c maximum earth-fault current onunderground systems is 1,000 A.The formula required to determine themaximum value of earthing resistance Rs atthe substation, to ensure that the LVwithstand voltage will not be exceeded, is:

Rs = Uw - Uo in ohms Im(see cases C and D in figure C7).WhereUw = the lowest standard value (in volts) of

short-term (5 s) withstand voltagefor the consumer's installation andappliances= 1.5 Uo + 750 V (IEC 644 (1991))

Uo = phase to neutral voltage (in volts) at theconsumer's LV service position

Im = maximum earth-fault current on the HVsystem (in amps).

A third form of system earthing referred to asthe “IT” system in IEC 364 is commonly usedwhere continuity of supply is essential, e.g. inhospitals, continuous-process manufacturing,etc.The principle depends on taking a supplyfrom an unearthed source, usually atransformer, the secondary winding of whichis unearthed, or earthed through a highimpedance (u 1,000 ohms).In these cases, an insulation failure to earthin the low-voltage circuits supplied from thesecondary windings will result in zero ornegligible fault-current flow, which can beallowed to persist until it is convenient toshut-down the affected circuit to carry outrepair work.

Diagrams B, D and F of figure C7 show ITsystems in which resistors (of approximately1,000 ohms) are included in the neutral-earthing lead.If however, these resistors were removed, sothat the system is unearthed, the followingnotes apply.Diagram B. All phase wires and the neutralconductor are “floating” with respect to earth,to which they are “connected” via the(normally very high) insulation resistancesand (very small) capacitances between thelive conductors and earthed metal (conduits,etc.).

Assuming perfect insulation, all LV phase andneutral conductors will be raised byelectrostatic induction to a potentialapproaching that of the equipotentialconductors.In practice, it is more likely, because of thenumerous earth-leakage paths of all liveconductors in a number of installations actingin parallel, that the system will behavesimilarly to the case where a neutral earthingresistor is present, i.e. all conductors will beraised to the potential of the substation earth.In these cases, the overvoltage stresses onthe LV insulation are small or non-existent.Diagrams D and F. In these cases, the highpotential of the substation (S/S) earthingsystem acts on the isolated LV phase andneutral conductors:c through the capacitance between the LVwindings of the transformer and thetransformer tank,c through capacitance between theequipotential conductors in the S/S andthe cores of LV distribution cables leavingthe S/S,c through current leakage paths in theinsulation, in each case.At positions outside the area of influence ofthe S/S earthing, system capacitances existbetween the conductors and earth at zeropotential (capacitances between cores areirrelevant - all cores being raised to the samepotential).The result is essentially a capacitive voltagedivider, where each “capacitor” is shunted by(leakage path) resistances.In general, LV cable and installation wiringcapacitances to earth are much larger, andthe insulation resistances to earth are muchsmaller than those of the correspondingparameters at the S/S, so that most of thevoltage stresses appear at the substationbetween the transformer tank and the LVwinding.The rise in potential at consumers’installations is not likely therefore to be aproblem where the HV earth-fault currentlevel is restricted as previously mentioned.All IT-earthed transformers, whether theneutral point is isolated or earthed through ahigh impedance, are routinely provided withan overvoltage limiting device which willautomatically connect the neutral pointdirectly to earth if an overvoltage conditionapproaches the insulation-withstand level ofthe LV system.In addition to the possibilities mentionedabove, several other ways in which theseovervoltages can occur are described inClause 3.1.


CThis kind of earth-fault is very rare, and whenit does occur is quickly detected and clearedby the automatic tripping of a circuit breakerin a properly designed and constructedinstallation.Safety in situations of elevated potentialsdepends entirely on the provision of properlyarranged equipotential areas, the basis ofwhich is generally in the form of a wide-meshed grid of interconnected bare copperconductors connected to vertically-drivencopper-clad* steel rods.The equipotential criterion to be respected isthat which is mentioned in Chapter G dealingwith protection against electric shock byindirect contact, namely: that the potentialbetween any two exposed metal parts whichcan be touched simultaneously by any partsof the body must never, under anycircumstances, exceed 50 V in dryconditions, or 25 V in wet conditions.Special care should be taken at theboundaries of equipotential areas to avoidsteep potential gradients on the surface of theground which give rise to dangerous “steppotentials”.This question is closely related to the safeearthing of boundary fences and is furtherdiscussed in Sub-clause 3.1 and inAppendix C2.* Copper is cathodic to most other metals and thereforeresists corrosion.

1.2 different HV service connections

According to the type of high-voltage network,the following supply arrangements arecommonly adopted.

single-line service

The substation is supplied by a single circuittee-off from a HV distributor (cable or line).In general, the HV service is connected into apanel containing a load-break/isolating switchwith series protective fuses and earthingswitches, as shown in figure C8.In some countries a pole-mountedtransformer with no HV switchgear or fuses(at the pole) constitutes the “substation”. Upto transformer ratings of 160 kVA this type ofHV service is very common in rural areas.Protection and switching devices are remotefrom the transformer, and generally control amain overhead-line, from which a number ofthese elementary service lines are tapped.

overhead line

fig. C8: single-line service.


C

ring-main principle

Ring-main units (RMU) are normallyconnected to form a HV ring main* orinterconnector-distributor*, such that the RMUbusbars carry the full ring-main orinterconnector current (figure C9).The RMU consists of three compartments,integrated to form a single assembly, viz:c 2 incoming compartments, each containinga load-break/isolating switch and a circuitearthing switch,c 1 outgoing and general protectioncompartment, containing a load-break switchand HV fuses, or a combined load-break/fuseswitch, or a circuit breaker and isolatingswitch, together with a circuit-earthing switchin each case.All load-break switches and earthing switchesare fully rated for short-circuit current-makingduty.This arrangement provides the user with atwo-source supply, thereby reducingconsiderably any interruption of service dueto system faults or operational manœuvresby the supply authority, etc.The main application for RMUs is in public-supply HV underground-cable networks inurban areas.* A ring main is a continuous distributor in the form of aclosed loop, which originates and terminates on one set ofbusbars. Each end of the loop is controlled by its own circuitbreaker. In order to improve operational flexibility thebusbars are often divided into two sections by a normally-closed bus-section circuit breaker, and each end of the ringis connected to a different section.An interconnector is a continuous untapped feederconnecting the busbars of two substations. Each end of theinterconnector is usually controlled by a circuit beaker.An interconnector-distributor is an interconnector whichsupplies one or more distribution substations along itslength.

parallel feeders

Where a HV supply connection to two lines orcables originating from the same busbar of asubstation is possible, a similar HVswitchboard to that of a RMU is commonlyused (figure C10).The main operational difference between thisarrangement and that of a RMU is that thetwo incoming panels are mutually interlocked,such that one incoming switch only can beclosed at a time, i.e. its closure prevents theclosure of the other.On the loss of power supply, the closedincoming switch must be opened and the(formerly open) switch can then be closed.The sequence may be carried out manuallyor automatically.This type of switchboard is used particularlyin networks of high-load density and inrapidly-expanding urban areas supplied byHV underground cable systems.

fig. C9: ring-main service.

undergroundcable ring main

fig. C10: duplicated supply service.

paralleled underground-cable distributors


1.2 different HV service connections (continued)


C

1.3 some operational aspects of HV distribution networks

overhead lines

High winds, ice formation, etc., can cause theconductors of overhead lines to touch eachother, thereby causing a momentary (i.e. notpermanent) short-circuit fault.Insulation failure due to broken ceramic orglass insulators, caused by air-borne debris;careless use of shot-guns, etc., or again,heavily polluted insulator surfaces, can resultin a short-circuit to earth.Many of these faults are self-clearing. Forexample, in dry conditions, broken insulatorscan very often remain in service undetected,but are likely to flashover to earth (e.g. to ametal supporting structure) during arainstorm. Moreover, polluted surfacesgenerally cause a flashover to earth only indamp conditions.The passage of fault current almost invariablytakes the form of an electric arc, the intenseheat of which dries the current path, and tosome extent, re-establishes its insulatingproperties. In the meantime, protectivedevices have usually operated to clear thefault, i.e. fuses have blown or a circuitbreaker has tripped.Experience has shown that in the largemajority of cases, restoration of supply byreplacing fuses or by re-closing a circuitbreaker will be successful.For this reason it has been possible toconsiderably improve the continuity of serviceon HV overhead-line distribution networks bythe application of automatic circuit breakerreclosing schemes at the origin of the circuitsconcerned.These automatic schemes permit a numberof reclosing operations if a first attempt fails,with adjustable time delays betweensuccessive attempts (to allow de-ionization ofthe air at the fault) before a final lock-out ofthe circuit breaker occurs, after all (generallythree) attempts fail.

Other improvements in service continuity areachieved by the use of remotely-controlledsection switches and by automatic isolatingswitches which operate in conjunction with anauto-reclosing circuit breaker.This last scheme is exemplified by the finalsequence shown in figure C11, where theisolating switch is referred to as IACT*(voltage-drop-operated outdoor switch).The principle is as follows:If, after two reclosing attempts, the circuitbreaker trips, the fault is assumed to bepermanent, and, while the distributor is dead,the IACT opens to isolate a section of thenetwork, before the third (and final) reclosuretakes place. There are then two possibilities:1) the fault is on the section which is isolatedby IACT, and supply is restored to thoseconsumers connected to the remainingsection, or2) the fault is on the section upstream ofIACT and the circuit breaker will trip and lockout.The IACT scheme, therefore, provides thepossibility of restoration of supplies to someconsumers in the event of a permanent fault.While these measures have greatly improvedthe reliability of supplies from HV overheadline systems, the consumers must, whereconsidered necessary, make their ownarrangements to counter the effects ofmomentary interruptions to supply (betweenreclosures), for example:c uninterruptible standby emergency power,c lighting that requires no cooling downbefore re-striking.(See Chapter F section 2)* Interrupteur Aérien à ouverture dans le Creux de Tension(used by EDF, the French supply authority).

If

In

Io

O1 RR O2

fault0.3s 0.4s

15 to 30s

SR O3

permanent fault

1-cycle RR + 1SR

fig. C11: automatic reclosing cycles of a circuit breaker controlling a radial HV distributor.

2-cycle 2SRa-fault on main distributor

If

In

Io

O1 RR O2

0.3s 0.4s

15 to 30s

SR1 O3

permanent fault0.4s

15 to 30s

SR2 O4

0.45sfault

b-fault on section supplied through IACT

If

In

Io

O1 RR O2

0.3s 0.4s

15 to 30s

SR1 O3

0.4s

15 to 30sSR2

opening of IACTfault

O = circuit breaker opening / RR = rapid reclosing / SR = slow reclosing / In = normal load current / If = fault current / I0 = zero current


C


1.3 some operational aspects of HV distribution networks (continued)

underground cable networks

Faults on underground cable networks aresometimes the result of carelessworkmanship by cable jointers or by cable-laying contractors, etc., but are morecommonly due to damage from tools such aspick-axes, pneumatic drills and trenchexcavating machines, and so on, used byother utilities.Insulation failures sometimes occur in cable-terminating boxes due to overvoltage,particularly at points in a HV system where anoverhead line is connected to anunderground cable. The overvoltage in sucha case is generally of atmospheric origin, andelectromagnetic-wave reflection effects at thejoint box (where the natural impedance of thecircuit changes abruptly) can result inoverstressing of the cable-box insulation tothe point of failure. Overvoltage protectiondevices, such as lightning arresters, arefrequently installed at these locations.Faults occurring in cable networks are lessfrequent than those on overhead (O/H) linesystems, but are almost invariably permanentfaults, which require more time for localizationand repair than those on O/H lines.Where a cable fault occurs on a ring main,supplies can be quickly restored to allconsumers when the faulty section of cablehas been determined.If, however, the fault occurs on a radialdistributor, the delay in locating the fault andcarrying out repair work can amount toseveral hours, and will affect all consumersdownstream of the fault position.In any case, if supply continuity is essentialon all, or part of, an installation, a standbysource must be provided. Standby powerequipment is described in Chapter Fsection 2.1.

remote control of HV networks

Remote control of HV circuit breakers andswitchgear, and tapchangers, etc. from acentral control room is possible, while similarcontrol facilities are also available from theconsole of a mobile control centre.

centralized remote control, based onSCADA (Supervisory Control AndData Acquisition) systems and recentdevelopments in IT (InformationTechnology) techniques, is becomingmore and more common in countriesin which the complexity of highly-interconnected systems justifies theexpenditure.


C

2. consumers HV substations

Large consumers of electricity are invariablysupplied at HV.On LV systems operating at 120/208 V(3-phase 4-wires), a load of 50 kVA might beconsidered to be “large”, while on a240/415 V 3-phase system a “large”consumer could have a load in excess of100 kVA. Both systems of LV distribution arecommon in many parts of the world.As a matter of interest, the IEC recommendsa “world” standard of 230/400 V for 3-phase4-wire systems. This is a compromise leveland will allow existing systems which operateat 220/380 V and at 240/415 V, or close tothese values, to comply with the proposedstandard simply by adjusting the off-circuittapping switches of standard distributiontransformers.The distance over which the load has to betransmitted is a further factor in consideringan HV or LV service. Services to small butisolated rural consumers are obviousexamples.The decision of a HV or LV supply willdepend on local circumstances andconsiderations such as those mentionedabove, and will generally be imposed by thepower-supply authority for the districtconcerned.When a decision to supply power at HV hasbeen made, there are two widely-followedmethods of proceeding.

(1) The power-supplier constructs a standardsubstation close to the consumer’s premises,but the HV/LV transformer(s) is (are) locatedin transformer chamber(s) inside thepremises, close to the load centre.(2) The consumer constructs and equips hisown substation on his own premises, towhich the power supplier makes the HVconnection.In method (1) the power supplier owns thesubstation, the cable(s) to the transformer(s),the transformer(s) and the transformerchamber(s), to which he has unrestrictedaccess.The transformer chamber(s) is (are)constructed by the consumer (to plans andregulations provided by the supplier) andinclude plinths, oil drains, fire walls andceilings, ventilation, lighting, and earthingsystems, all to be approved by the supplyauthority.The tariff structure will cover an agreed partof the expenditure required to provide theservice.Whichever procedure is followed, the sameprinciples apply in the conception andrealization of the project. The following notesrefer to procedure (2).

2.1 procedures for the establishment of a new substation

the consumer must provide certaindata to the power-supplyorganization at the earliest stageof the project.

preliminary information

Before any negotiations or discussions canbe initiated with the supply authorities, thefollowing basic elements must beestablished:c maximum anticipated power (kVA)demandDetermination of this parameter is describedin Chapter B, and must take into account thepossibility of future additional loadrequirements. Factors to evaluate at thisstage are:v the utilization factor (ku),v the simultaneity factor (ks),c layout plans and elevations showinglocation of proposed substationPlans should indicate clearly the means ofaccess to the proposed substation, with

dimensions of possible restrictions, e.g.entrances corridors and ceiling height,together with possible load (weight) bearinglimits, and so on, keeping in mind that:v the power-supply personnel must have freeand unrestricted access to the HV equipmentin the substation at all times,v only qualified and authorized consumer’spersonnel are allowed access to thesubstation.c degree of supply continuity requiredThe consumer must estimate theconsequences of a supply failure in termsof its duration,v loss of production,v safety of personnel and equipment.

project studies

From the information provided by theconsumer, the power-supplier must indicate:c the type of power supply proposedand define:v the kind of power-supply system: overhead-line or underground-cable network,v service connection details: single-lineservice, ring-main installation, or parallelfeeders, etc.,v power (kVA) limit and fault current level.

c the nominal voltage and rated voltage(Highest voltage for equipment)Existing or future, depending on thedevelopment of the system.c metering detailswhich define:v the cost of connection to the powernetwork,v tariff details (consumption and standingcharges).

the power-supply organization mustgive specific information to theprospective consumer.


C

2. consumers HV substations (continued)

2.1 procedures for the establishment of a new substation (continued)

the power-supply organization mustgive official approval of theequipment to be installed in thesubstation, and of proposed methodsof installation.

implementation

Before any installation work is started, theofficial agreement of the power-supplier mustbe obtained. The request for approval mustinclude the following information, largelybased on the preliminary exchanges notedabove:c location of the proposed substation,c one-line diagram of power circuits andconnections, together with earthing-circuitproposals,

c full details of electrical equipment to beinstalled, including performancecharacteristics,c layout of equipment and provision formetering components,c arrangements for power-factorimprovement if eventually required,c arrangements provided for emergencystandby power plant (HV or LV) if eventuallyrequired.

commissioning

Commissioning tests must be successfullycompleted before authority is given toenergize the installation from the power-supply system. The verification tests includethe following:c measurement of earth-electroderesistances,c continuity of all equipotential earth-andsafety bonding conductors,c inspection and testing of all HVcomponents,c insulation checks of HV equipment,c dielectric strength test of transformer oil(and switchgear oil if appropriate),c inspection and testing of the LV installationin the substation,c checks on all interlocks (mechanical keyand electrical) and on all automaticsequences,c checks on correct protective-relay operationand settings.It is also imperative to check that allequipment is provided, such that any properlyexecuted operational manœuvre can becarried out in complete safety.On receipt of the certificate of conformity:c personnel of the power-supply authority willenergize the HV equipment and check forcorrect operation of the metering,c the installation contractor is responsible fortesting and connection of the LV installation.

When finally the substation is operational:c the substation and all equipment belongs tothe consumer,c the power-supply authority has operationalcontrol over all HV switchgear in thesubstation, e.g. the two incoming load-breakswitches and the transformer HV switch (orCB) in the case of a RMU, together with allassociated HV earthing switches,c the power-supply personnel hasunrestricted access to the HV equipment,c the consumer has independent control ofthe HV switch (or CB) of the transformer(s)only,c the consumer is responsible for themaintenance of all substation equipment, andmust request the power-supply authority toisolate and earth the switchgear to allowmaintenance work to proceed. The powersupplier must issue a signed permit-to-workto the consumers maintenance personnel,together with keys of locked-off isolators, etc.at which the isolation has been carried out.

after testing and checking of theinstallation by an independent testauthority, a certificate is grantedwhich permits the substation to beput into service.


C

3. substation protection schemes

The subject of protection in the electrical-power industry is vast: it covers all aspects ofsafety for personnel, and protection againstdamage or destruction of property, plant, andequipment.These different aspects of protection can bebroadly classified according to the followingobjectives:c protection of personnel and animals againstthe dangers of overvoltages and electricshock, fire, explosions, and toxic gases, etc.,c protection of the plant, equipment andcomponents of a power system against thestresses of short-circuit faults, atmosphericsurges (lightning) and power-systeminstability (loss of synchronism) etc.,c protection of personnel and plant from thedangers of incorrect power-system operation,by the use of electrical and mechanicalinterlocking. All classes of switchgear(including, for example, tap-position selectorswitches on transformers, and so on...) havewell-defined operating limits. This means thatthe order in which the different kinds ofswitching device can be safely closed oropened is vitally important. Interlocking keysand analogous electrical control circuits arefrequently used to ensure strict compliancewith correct operating sequences.It is beyond the scope of a guide to describein full technical detail the numerous schemesof protection available to power-systemsengineers, but it is hoped that the followingsections will prove to be useful through adiscussion of general principles. While someof the protective devices mentioned are ofuniversal application, descriptions generallywill be confined to those in common use onHV and LV systems only, as defined inSub-clause 1.1 of this Chapter. Where sometechnical explanation is necessary to simplifyan understanding of the text, reference ismade to a related Appendix.

3.1 protection against electric shocks and overvoltages

protection against electric shocksand overvoltages is closely related tothe achievement of efficient (lowresistance) earthing and effectiveapplication of the principles ofequipotential environments.

protection against electric

shocks

Protective measures against electric shockare based on two common dangers:c contact with an active conductor, i.e. whichis alive with respect to earth in normalcircumstances. This is referred to as a “directcontact” hazard,c contact with a conductive part of anapparatus which is normally dead, but whichhas become alive due to insulation failure inthe apparatus. This is referred to as an“indirect contact” hazard.It may be noted that a third type of shockhazard can exist in the proximity of HV or LV(or mixed) earth electrodes which are passingearth-fault currents. This hazard is due topotential gradients on the surface of theground and is referred to as a “step-voltage”hazard; shock current enters one foot andleaves by the other foot, and is particulardangerous for four-legged animals. Avariation of this danger, known as a “touchvoltage” hazard can occur, for instance,where an earthed metal fence is situated inan area in which potential gradients exist.Touching the fence would cause shockcurrent to pass through the hand and bothfeet (Appendix C2).

Animals with a relatively long front-to-hindlegs span are particularly sensitive to step-voltage hazards and cattle have been killedby the potential gradients caused by a lowvoltage (240/415 V) neutral earth electrode ofinsufficiently low resistance. Potentialgradients on the surface of the ground can bereduced to safe values by measures such asthose shown in Appendix C2.Potential-gradient problems of the kindmentioned above are not normallyencountered in electrical installations ofbuildings, providing that equipotentialconductors properly bond all exposed metalparts of equipment and all extraneous metal(i.e. not part of an electrical apparatus or theinstallation - for example structural steel-work, etc.) to the protective-earthingconductor.


C

3. substation protection schemes (continued)

3.1 protection against electric shocks and overvoltages (continued)

Direct-contact protectionThe main form of protection against directcontact hazards is to contain all live parts inhousings of insulating material, by placing outof reach (behind insulated barriers or at thetop of poles) or by means of obstacles.Where insulated live parts are housed in ametal envelope, for example transformers,electric motors and many domesticappliances, the metal envelope is connectedto the installation protective earthing system.For LV appliances this is achieved throughthe third pin of a 3-pin plug and socket. Totalor even partial failure of insulation to themetal, can (depending on the ratio of theresistance of the leakage path through theinsulation, to the resistance from the metalenvelope to earth) raise the voltage of theenvelope to a dangerous level.

Indirect-contact protectionA person touching the metal envelope of anapparatus of which the insulation is faulty, asdescribed above, is said to be making anindirect contact.An indirect contact is characterized by thefact that a current path to earth exists(through the protective earthing (PE)conductor) in parallel with the shock currentthrough the person concerned.c case of fault on L.V. system.Extensive tests have shown that, providingthe potential of the metal envelope is notgreater than 50 V* with respect to earth, or toany conductive material within reachingdistance, no danger exists.c indirect-contact hazard in the case of a HVfault.If the insulation failure in an apparatus isbetween a HV conductor and the metalenvelope, it is not generally possible to limitthe rise of voltage of the envelope to 50 Vor less, simply by reducing the earthingresistance to a low value.The solution in this case is to create anequipotential situation, as described inSub-clause 1.1 “Earthing connections”.

* in dry locations, 25 V in wet locations(bathrooms, etc.).

in the case of a HV fault to a metallicenclosure, it may not be possible tolimit the touch voltage to the safevalue of 50 V*. The solution is tocreate an equipotent-situation asdescribed in Sub-clause 1.1"Earthing connections".

protection against overvoltages

The situation mentioned immediately above,describing an indirect-contact hazardresulting from faulty HV insulation, is oneof a number of ways in which an abnormalovervoltage condition can occur.Methods of eliminating danger to personnel insuch a case are described in Sub-clause 1.1.Other situations which can causeovervoltages to occur on HV and LV systemsinclude:c surges of atmospheric origin,c a short-circuit earth fault on an unearthed(or high-impedance earthed) 3-phase system,c ferro-resonance,c energization of capacitor banks,c circuit breaker opening or fuse melting tobreak short-circuit current.Overvoltages created by the causes listedabove can be divided according tocharacteristies such as:c duration: permanent, temporary, transient,c frequency: industrial frequency, harmonicsof industrial frequency, high frequency,unidirectional surges.

Overvoltages of atmospheric originProtection against this kind of danger must beprovided when a substation is supplieddirectly from an overhead-line system. Themost common protective device used atpresent is a non-linear resistor-type oflightning arrester, which is connected (one foreach phase) between a phase conductor andthe substation earthing system, as close tothe point of entry into the substation aspossible.For consumers' substations, this protection isachieved by:c lightning arresters (one per phaseconductor, which are sometimes connected inseries with a device for automatic tripping of acircuit breaker) (see Chapter L) and/or byc the reduction of the substation-earthingresistance to the lowest possible value toavoid (as far as possible) a breakdown of LVinsulation due to the rise in potential of theearthing system when discharging the surgecurrent.Where it is advisable to protect a substationagainst direct strokes, lightning-dischargeelectrodes (Franklin type) and shield wiresshould be installed and connected to thesubstation earthing system.It may be noted that, at the voltage levelsbeing considered (i 35 kV), switching surgesare generally less severe than lightningsurges, and so devices which are suitable forsatisfactory lightning protection are adequateto protect against overvoltages dueto switching surges.


CEarth faults on IT-earthed systemsIn normal conditions the phase conductors ofa 3-phase IT system are all approximately atphase volts with respect to earth. The exactvalues depend on the capacitance andinsulation resistance of each conductor toearth. On an unfaulted system theseparameters are sensibly equal in all threephases so that the vector relationship ofphase voltages will be as shown below infigure C12, and the neutral point of thetransformer secondary winding will be atapproximately zero volts with respect toearth.A short-circuit to earth on one phase willchange the values of phase conductorvoltages with respect to earth; the phase-to-phase voltage values and their phasedisplacement relationships however, willremain unchanged. This latter feature,

together with the fact that current passingthrough the earth-fault path will be too smallto constitute a hazard, are the reasons foradopting the IT system, where supplycontinuity must be maintained even in “first-fault” conditions.With one phase short-circuited to earth, andthe neutral point of the transformer isolated:c the neutral point will rise to phase voltsabove earth,c the faulty phase conductor will be at zerovolts with respect to earth,c the other two phases will rise to etimesthe phase voltage, with respect to earth.As indicated above, this is a 50 Hz (or 60 Hz)stable (i.e. permanent) condition, andtransformers, cabling and all appliances mustbe suitably insulated with respect to earth,when used on IT systems.

normal voltages and capacitive / resistive currents

V1

I(C+R)3 V2V3

I(C+R)1 I(C+R)2

<90°

unearthed secondary winding of power transformer

fault current normally restricted to several milli-amps depending on the size of the installation

N

1

2

3

insulationresistance

conductorcapacitance

voltage conditions and current flowing in an earth fault on an IT system

V1

V2

√3 I(C+R)1VNE

V3 = 0

√3 I(C+R)2

3 I(C+R)

fig. C12: earth fault on IT-earthed systems.


C


3.1 protection against electric shocks and overvoltages (continued)

Ferro-resonanceFerro-resonance is a spontaneous conditionwhich occurs due to a complex interactionbetween intrinsic power-system capacitancesand the non-linear voltage-dependentinductances of transformers or reactors,chokes, etc. when their magnetic circuits arein a highly-saturated state (generally due toan abnormal system disturbance).The resonant condition may be at anyfrequency, and can be a parallel or seriesresonance, not corresponding exactly to theclassical formulae for LC resonant circuits(which are based on assumed linearity of theLC components). Moreover, resonance mayoccur on one or two phases only of a 3-phasesystem.All types of transformer can be affected,including instrument voltage transformers;the capacitor-type VT (not normally usedat the HV levels considered in this guide) isespecially prone to sub-harmonic resonance(1/3 of the fundamental frequency).Electro-magnetic VTs (which are verycommonly used at HV levels covered by thisguide) counter the possibility of resonanceby:c designing the transformer cores to operateat low levels of flux density,c incorporating damping resistors in thetransformer secondary or tertiary circuits.Apart from the obvious problems presentedby false signals given by instrumenttransformers, permanent overvoltageconditions can be established.Unless the precautions mentioned above aretaken, the following situation may arise (andoften did in the past, before the phenomenonwas identified). The problem concernsIT-earthed systems, in which the potential ofthe neutral point becomes displaced (fromapprox. earth potential) with the result thatexcessive values of phase voltage withrespect to earth occur on two phases, asshown in the vector diagram of figure C13.The condition is due to the saturation of two(of the three) single-phase cores of a voltagetransformer, the windings of which areconnected between phase and earth, asshown in figure C14, and may be provokedby a transitory* overvoltage condition such asthat described above and shown in figureC12.The overvoltage saturates the two single-phase VT cores, which then present a (non-linear, but average) inductance which is muchlower than its normal value. The parallelcombination of phase-to-earth capacitanceand phase-to-earth inductance which, undernormal conditions, behaves overall as acapacitance K (because the capacitivereactance < inductive reactance) suddenlychanges character to that of an inductance.The two states, viz: before saturation andduring saturation are shown in figure C14 (a)and (b) respectively.In C14 (a1) the three capacitances and thethree inductances each form an independentbalanced 3-phase star-connected group, i.e.there is no exchange of current betweenthem.* for example, a bird causing a brief short-circuit to earth,and falling clear of the line, or wind-blown debris, etc.

V1N

V2N

E

V2E

N

V3N

V1E

V3E

fig. C13: vector diagram of a displacedneutral due to ferro-resonance at 50 Hz.


CIn C14 (b2) however, it will be seen thatthe two inductances, together with onecapacitance, constitute an unbalanced3-phase group, the star point of whichis the earth.It is clear that an unbalanced 3-phase load ona 3-wire system will displace the “floating”neutral point of the source. A simplecalculation in Appendix C3 shows how thevector diagram of figure C13 was determined.

Note: it should be appreciated that the vectorrepresentation shown in figure C13 gives anapproximate, but qualitatively useful, pictureof the actual phenomenon, since harmonicvoltages and currents are also present.However, in the 50 Hz (or 60 Hz) resonantcondition, the power-frequency values arepredominant. The circuit behaviour therefore,is mainly governed by these power-frequencyquantities, and justifies an approximaterepresentation by vectors. Fieldmeasurements have confirmed the validity ofsuch a representation.For interested readers, further information onferroresonance can be found in CahierTechnique No. 31 : "Ferroresonance"published by Merlin Gerin.

(a) circuits in normal operation

K

(b1)

=

H < L XH > XK at a resonant frequencyHH

(b) circuits with phase 1 and phase 2 VT cores saturated

K < Cat normal system frequency

==(a2)

K K K

source

N

1

23

(a1)

L L L C C C

source

(b2)

N

fig. C14: equivalent circuits for ferro-resonant condition.


C


general

The circuits and equipment in a substationmust be protected so that excessive currentsand/or voltages are rapidly removed from thesystem before causing danger, damage ordestruction. All equipments normally used inpower-system installations have(standardized) short-time withstand ratings forovercurrent and overvoltage conditions, andthe role of protective schemes is to ensurethat these withstand limits can never beexceeded. In general, this means that faultconditions must be cleared as fast aspossible within the limits set byconsiderations of the highest attainablereliability.Overcurrents due to overloading can normallybe tolerated for longer periods than those ofshort-circuits and some protective devicesare designed to operate with increasingspeed as the degree of overloading increases(i.e. an inverse-time/current characteristic).In addition to the protection againstovervoltages mentioned in section 3.1,electrical protection is routinely providedagainst the following abnormal conditions:c overloading (i.e. excessive currents not dueto faults),c transformer faults,c short-circuit faults between phases,c short-circuit faults to earth, and iscommonly realized by:c a circuit breaker downstream of thetransformer,c detection and trip-initiating devices whichare integral parts of the transformer,c a circuit breaker or fuses (with or without anassociated load-break switch) upstream ofthe transformer.The choice and sophistication of theprotective schemes will depend on thecharacteristics of the substation, and arediscussed later.Protective devices upstream of thetransformer must be co-ordinated withdownstream devices, as noted in Chapter H2,Sub-clause 4.6.

3.2 electrical protection

overcurrents due to overloading orto short-circuit faults (betweenphases and/or to earth) are detectedby protective devices up-stream anddown-stream of the powertransformer(s). These devicescause the faulty circuit to be cut-offelectrically from the power supply.The devices may be:c fuses which clear the faulty circuitdirectly, or together with amechanical tripping attachment,which opens an associated three-phase load-break switch,c direct-acting tripping coils whichform part of a LV circuit breaker, andare operated by the fault (oroverload) current passing throughthem,c relays which act indirectly such as:v electrical relays supplied fromcurrent and/or voltage transformers,v pressure-operated relays(pressostats),v temperature-operated relays(thermostats),v gas-detection relays (Buchholz,etc.),v oil-surge operated relays.

overload protectionOverloading is frequently due to the co-incidental demand of a number of smallloads, or to an increase in the apparentpower (kVA) demand of an installation, due toexpansion of an enterprise, with consequentbuilding extensions, and so on.-Loadincreases raise the temperature of the circuitconductors concerned, together with that ofthe transformer.When the temperature exceeds the normaldesign limits of the equipment involved, thedeterioration rate (ageing) of the insulationmaterials is increased, and the working life ofthe equipment is correspondingly reduced.Overload protection devices are usuallylocated downstream of the transformer inconsumer-type substations, but arecommonly provided on the upstream side inpublic-supply substations.


Ctransformer protection

OverloadsThe protection against overloading of atransformer is provided by a time-delayedoverload relay (either a thermal bi-metal stripdevice or an electronic device) which acts totrip the downstream-side circuit breaker.The time delay inherent in this relay ensuresthat the transformer will not be unnecessarilytripped for overloads of short duration. Otheralternatives are:c for pole-mounted transformers “thermal-image” relays are frequently used. Suchrelays artificially simulate the temperature ofthe transformer windings with an accuracywhich is sufficient to safeguard the insulation,c dry-type transformers use heat sensorsembedded in the hottest part of the windingsinsulation for alarm and/or tripping, whilec larger oil-immersed transformers frequentlyhave thermostats with two settings, one foralarm purposes and the other for tripping.Internal faultsThe protection of transformers bytransformer-mounted devices, against theeffects of internal faults, is provided ontransformers which are fitted with air-breathing conservator tanks (see figureC15), by the classical Buchholz mechanicalrelay. These relays can detect a slowaccumulation of gases which results from thearcing of incipient faults in the windinginsulation or from the ingress of air due to anoil leak. This first level of detection generallygives an alarm, but if the conditiondeteriorates further, a second level ofdetection will trip the upstream circuit breaker.An oil-surge detection feature of the Buchholzrelay will trip the upstream circuit breaker"instantaneously" if a surge of oil occurs inthe pipe connecting the main tank with theconservator tank.Such a surge can only occur due to thedisplacement of oil caused by a rapidly-formed bubble of gas, generated by an arc ofshort-circuit current under the oil.All transformers are fitted with some kindover-pressure relief device, which limits themaximum pressure to a value well below thatat which the transformer tank will rupture.By specially designing the cooling-oil radiatorelements to perform a concertina action,“totally filled” types of transformer as large as10 MVA are now currently available.Expansion of the oil is accommodated withoutan excessive rise in pressure by the “bellows”effect of the radiator elements. A fulldescription of these transformers is given inSub-clause 4.4 (see figure C16).Evidently the Buchholz devices mentionedabove cannot be applied to this design; amodern counterpart has been developedhowever, which measures:c the accumulation of gas,c overpressure,c overtemperature,the first two conditions trip the upstreamcircuit breaker, and the third condition tripsthe downstream circuit breaker of thetransformer.The device, referred to as a "DGPT" unit(Detection of Gas, Pressure andTemperature) is mentioned further in 4.4,under "Liquid-filled transformers".

fig. C15: transformer with conservatortank.

fig. C16: total-fill transformer.


C


3.2 electrical protection (continued)

fig. C17: protection against earth fault onthe HV winding.

protection against short-circuits

Short-circuits may occur between phaseconductors, between a phase conductor andearth, or in any combination of theseconditions on the three phases. The(extremely unlikely) occurrence of a short-circuit fault between the high-voltagewindings and the low-voltage windings willconstitute a short-circuit-to-earth fault on theHV winding if the secondary winding isearthed, which is generally the case.Unearthed star-connected LV secondarywindings of IT-system transformers have anovervoltage device which will operate in thesecircumstances to connect the LV neutral pointof the transformer directly to earth.Earth faults on the HV winding present aparticular danger to personnel, due to theTransferred Potential hazard mentioned inSub-clause 1.1: “Earthing connections”.For this reason, high-speed sensitive earth-fault protection is standard on the HV side ofpower transformers in many public-supplydistribution and consumer-type substations.The scheme is shown below in figure C17and can be applied to transformers havingdelta or unearthed-star primary windings.This arrangement is called a “RestrictedEarth-Fault" (REF) protection because it willdetect earth faults only on the HV windings oron the circuit downstream of the CTs (currenttransformers) to the winding terminals.The advantages of the scheme include:c simplicity and low cost,c instantaneous operation,c high sensitivity,c virtual elimination of the dangers ofTransferred Potential (because of itsinstantaneous operation),c no problems of coordination withdownstream protection; LV earth faultsappear as phase/phase faults on the HV sideof the transformer, thereby being undetectedby the REF relay (see fig. AC1-2(c) ofAppendix C1).Again, in public-supply systems in general,there is no CB on the LV side, simply a load-break switch. Overcurrent protective relays(2 only) are connected in series with the REFcurrent transformers, as shown dotted infigure C17 (see note). These overcurrentrelays afford protection against overloadingand short-circuit faults downstream of theCTs, but must be carefully co-ordinated withLV overcurrent protective devices.

Note: where short-circuit fault levels are low,it is recommended to use 3 overcurrentrelays (rather than 2) since on delta/startransformers a phase/phase short-circuit atLV gives a 2:1:1 fault-current distribution atHV (see fig. AC1-2(b) of Appendix C1).In view of the effectiveness of REF protectionagainst the hazards of transferred potentials,and its uncomplicated application, it isstrongly recommended to be included in anyprotection scheme which includes a HVcircuit breaker.

choice of protective devices

on the upstream side of the

transformer in a consumer-type

substation

In certain national standards*, the choice ismade according to two current values:c the reference current Ib, the value of whichwill be:v in the case of metering at low voltage: thenominal rated current of the transformer,v in the case of metering at high voltage: thesum of the nominal rated currents oftransformers and other HV plant (e.g. motors,etc.),c the minimum value of HV 3-phase short-circuit current at the installation.* There is no equivalent IEC standard.

N

3

2

11

2

3

HV LV

E/F relay


CWhen the reference current is less than 45 Aand there is only one transformer, theprotection may be by fuses or by a circuitbreaker.When the reference current is equal to orgreater than 45 A, or when there is more thanone transformer, the protection will be by acircuit breaker.The maximum IEC standard kVA ratings oftransformers corresponding to a HV full-loadcurrent not exceeding 45 A are given in thetable C18.

c protection by fusesThe relationships between the referencecurrent Ib, as defined above, the rated currentIn of the fuse, and the short-circuit current Icat the primary terminals of the transformer,are determined according to the nationalstandards previously referred to, as follows:v when the substation consists of a singleHV/LV transformer, the rated current In of thefuse must satisfy the following relationships:In > 1.4 Ib and In < Ic/6Where:In = rated current of the fuse,Ib = rated primary current of the transformer,Ic = the minimum current at the primary sideof the transformer when the secondary-winding terminals are short-circuited,v when the substation is supplied from anoverlead line, or when the installation issensitive to unbalanced-voltage conditions(for example three-phase motor loads), it isrecommended that the failure of a fusecauses all three phases to be cut off, byautomatic tripping of the HV load-breakswitch (i.e. a combined switch-fuse).Standard current ratings for fuses accordingto IEC 282-1 are listed in table C19.

table C18: power limits of transformerswith a maximum primary current notexceeding 45 A.

primary maximum IECvoltage standard ratings(kV) for transformersrated nominal (kVA)3.6 3 250

3.37.2 4.16

5.5 50066.6

12 10 80011

17.5 13.8 1,25015

24 20 1,60022

36 33 2,50040.5 36.5 3,150


C



supply voltages nominal transformer ratings(kV) (kVA)rated nominal 25 50 100 125 160 200 250 315 400 500 630 800 1,000 1,250 1,600 2,000 2,5003.6 3 16 25 40 50 50 63 80 80 100 125 160 200 250

3.3 16 25 40 50 50 63 80 80 100 125 160 200 2507.2 4.16 10 25 31.5 40 50 50 63 80 80 100 125 160 200 250

5.5 10 16 25 31.5 40 40 50 63 63 80 100 125 160 200 2506 10 16 25 31.5 31.5 40 50 50 63 80 80 100 125 160 200 2506.6 10 16 25 25 31.5 40 40 50 63 80 80 100 125 160 200 250

12 10 6.3 10 16 25 25 31.5 31.5 40 50 50 63 80 80 100 125 160 20011 6.3 10 16 25 25 25 31.5 31.5 40 50 63 63 80 100 125 160 200

17.5 13.8 6.3 6.3 10 16 25 25 25 31.5 31.5 40 50 63 63 80 100 160 16015 6.3 6.3 10 16 16 25 25 31.5 31.5 40 50 50 63 80 80 100 160

24 20 6.3 6.3 10 10 16 16 25 25 31.5 31.5 40 50 50 63 80 80 16022 6.3 6.3 10 10 10 16 25 25 25 31.5 31.5 40 50 50 63 80 160

36 33 6.3 6.3 6.3 6.3 6.3 6.3 16 16 16 16 16 31.5 31.5 40 50 63 8040.5 36.5 6.3 6.3 6.3 6.3 6.3 6.3 16 16 16 16 16 25 25 31.5 40 50 63

table C19: rated current (A) of HV fuses for transformer protection according to IEC 282-1.

It is strongly recommended that, following theoperation of a fuse (or fuses) to clear a faultor overload condition, all three fuses bereplaced, since it is possible that the fuse (orfuses) that had not operated may havedeteriorated, due to the passage of excessivecurrent during the disturbance.

c protection by circuit breakerWhen the substation is supplied through aHV circuit breaker, it will be a contractualcondition that no disturbance occurring withinthe installation shall cause the operation ofany protective relaying in the power-supplynetwork.To ensure that this condition can be compliedwith, the supply authority must specify thelongest times permissible for clearing thefollowing faults on the installation:c short-circuit fault between all 3 phases,c short-circuit fault between any 2 phases,c short-circuit fault of one phase to earth,c short-circuit fault of any 2 phases to earth.The maximum level of a 3-phase short-circuitat the installation was known at the outset ofthe project, in order to purchase adequately-rated equipment. To ensure correct operationof the protective devices, the minimum valueof 3-phase short-circuit current must also bestated by the supply authority.When planning the protection scheme for theinstallation, the general principle of co-ordination is that the circuit breaker closest tothe power source will have the longesttripping time. In the present case, this is theHV circuit breaker.This longest tripping time must not, however,exceed that given by the supply authority; aconstraint which generally is satisfied only byprotective relays at the HV circuit breaker, tosupplement the transformer-mountedprotective devices previously mentioned.As far as earth faults are concerned, there isno co-ordination problem, provided that thetransformer HV winding is a delta- orunearthed-star connection, since, as alreadymentioned, earth faults on the LV system willthen appear as phase-to-phase faults on theHV system. HV earth faults occurring in thesubstation can therefore be clearedinstantaneously by a REF scheme.Instantaneous tripping for phase-to-phaseshort-circuit faults occurring on the HV side of

the transformer can also be achieved verysimply by devices which are sometimesreferred to as “high-set” relays.The “high-set” principle depends on the factthat, if the current is sufficiently high tooperate the relay, then the short-circuit mustbe on the HV side of the transformer,because a short-circuit on the LV terminals orwindings of the transformer will not producesufficient current on the HV side to cause therelay to operate.The high-set relays (2 or 3 as noted in Sub-clause "protection against short-circuits") willeach be connected in series with one of theinverse-time/overcurrent relays, shown dottedin figure C17, and for distribution-typetransformers are generally set to operate at25 times the full-load current of thetransformer. By these simple meanstherefore, instantaneous clearance of short-circuit faults on the HV side of a transformercan be achieved, without affecting the co-ordination scheme for downstream protection.At periods of the lowest levels of short-circuitfault current, however, the high-set schememay not be sufficiently sensitive, i.e. thecurrent may not be high enough to operatethe relay (there is no similar problem withearth faults, the REF scheme being verysensitive).In extreme cases, where the differencebetween maximum and minimum fault levelsis very large, it may be necessary to provide adifferential-protection scheme for thetransformer.Differential-protection schemes compare thecurrents entering the primary windings withthose leaving the secondary windings (aftercorrection for current-level and phasechanges) and any significant difference willoperate the relay, which trips the circuitbreakers controlling the transformer.Such protection will provide adequatesensitivity with high-speed tripping, and willnot affect co-ordination of downstreamprotection.It may be noted that the high-speed relaysused for the REF, high-set and differentialprotection schemes are stabilized againstfalse operation due to CT saturation (forexample, when energizing the transformer).Overcurrent, REF, and high-set relays arecommonly contained in a single relay case.

no disturbance occurring within theinstallation shall cause the operationof any protective relaying in thepower-supply network.


C

HV fuses and LV circuit breakers.*In order to compare the two curves, the HVcurrents must be converted to the equivalentLV currents, or vice-versa.Figure C21 illustrates these requirements.Where an LV fuse-switch is used, similarseparation of the characteristic curves of theHV and LV fuses must be respected. Thisquestion is considered in Appendix C1(figure AC1-3).

choice of downstream protective

devices

The protective device (circuit breaker orfuse-switch)* downstream of the transformermust include and comply with the followingrequirements (IEC 364). The device must:c include an isolating switch (for theprotection of persons) of which the openswitch contacts are clearly visible,c have a current rating adequate for thetransformer concerned,c have a breaking-current rating whereappropriate*, adequate for the secondary3-phase short-circuit current,c have the correct number of poles accordingto the earthing scheme of the installation,

table C20: 3-phase short-circuit currents of typical distribution transformers.

discrimination (selectivity)

between the protective devices

on the upstream and

downstream sides of the

transformer

The consumer-type substation with LVmetering requires discriminative operationbetween the HV fuses and the LV circuitbreaker or fuses. The calibre of the HV fuseswill have been chosen according to thecharacteristics of the transformer.The tripping characteristics of the LV circuitbreaker must be such, that for an overload orshort-circuit condition downstream of itslocation, the breaker will trip sufficientlyquickly to ensure that the HV fuses will not beadversely affected by the passage ofovercurrent through them.The tripping performance curves for HV fusesand LV circuit breakers are given by graphsof time-to-operate against current passingthrough them. Both curves have the generalinverse-time/current form (with an abruptdiscontinuity in the CB curve at the currentvalue above which “instantaneous” trippingoccurs).These curves are shown typically infigure C21.c in order to achieve discrimination:v all parts of the fuse curve must be aboveand to the right of the CB curve,c in order to leave the fuses unaffected (i.e.undamaged):v all parts of the minimum pre-arcing fusecurve must be located to the right of the CBcurve by a factor of 1.35 or more (e.g. where,at time T, the CB curve passes through apoint corresponding to 100 A, the fuse curveat the same time T must pass through a pointcorresponding to 135 A, or more, and soon...) and,v all parts of the fuse curve must be abovethe CB curve by a factor of 2 or more (e.g.where, at a current level I the CB curvepasses through a point corresponding to1.5 seconds, the fuse curve at the samecurrent level I must pass through a pointcorresponding to 3 seconds, or more, etc.).The factors 1.35 and 2 are based on standardmaximum manufacturing tolerances for

transformer rated power (kVA) 50 100 160 250 315 400 500 630 800 1000 1250 1600 2000 2500transformer current Ir (A) 69 137 220 344 433 550 687 866 1100 1375 1718 2199 2749 3437oil-immersed transformer Isc (kA) Psc = 250 MVA 1.71 3.40 5.41 8.38 10.5 13.2 16.4 20.4 17.4 21.5 26.4 33.1 40.4 49.1

Psc = 500 MVA 1.71 3.42 5.45 8.49 10.7 13.5 16.8 21.0 17.9 22.2 27.5 34.8 43.0 52.9cast-resin transformer Isc (kA) Psc = 250 MVA 1.14 2.28 3.63 5.63 7.07 8.93 11.1 13.9 17.4 21.5 26.4 33.1 40.4 49.1

Psc = 500 MVA 1.14 2.28 3.65 5.68 7.14 9.04 11.3 14.1 17.9 22.2 27.5 34.8 43.0 52.9

D

C

time

AB

current

minimum pre-arcing time of HV fuse

circuit breaker tripping characteristic

B/A u 1.35 at any moment in timeD/C u 2 at any current value

v 4 poles for IT scheme with neutralconductor, TT and TNS,v 3 poles for IT scheme without neutralconductor, and for TNC.By way of an example, table C20 lists thenominal currents and corresponding short-circuit currents at the secondary terminals ofIEC-standard 20 / 0.4 kV transformers. Fromthese data it can be seen that the short-circuitimpedances are in the range of 4% (for the100 kVA transformer) to 6.9% (for the2,000 kVA transformer).* where no LV circuit breaker or fuse-switch is installed, anon-automatic LV load-break isolating switch must beprovided, and overload protection must be effected at HV.

Note: In the simple and widely used case,where a HV circuit breaker incorporates REF,high-set and inverse-time/overcurrent relaysas previously described, the only electricalprotection of the LV windings and the LVconnections from the transformer terminals tothe upstream terminals of the LV circuitbreaker is that provided by the HV inverse-time/overcurrent relays.* Merlin Gerin “catalogue distribution HT/MT 96” page G29.

fig. C21: discrimination between HV fuseoperation and LV circuit breaker tripping,for transformer protection.


C



There is no compelling need fordiscrimination between these HV relays andthe LV circuit breaker protection, since ashort-circuit on the upstream or downstreamside of the LV breaker would result in a totalloss of supply, in either case.In general, a 3-phase short-circuit at the LVterminals of a distribution transformer willcause a current of 14-25 times thetransformer full-load current to flow in the LVand HV circuits (at times of maximum short-circuit fault levels on the system).If the fault is a short-circuit of one phase toearth, immediately upstream of the LV circuitbreaker however, then this value will bereduced to approximately 8-14 times full-load

current on the HV side of the transformer andwill flow in two lines only. The tripping time ofthe inverse-time/overcurrent relays may, inthis case, be unacceptably long.The conventional solution to the problem is tomake these LV connections “fault-free” byenclosing the (generously-insulated)conductors in vermin-proof metal bus-ducts, amethod which, in view of the location (in anarea prohibited to all except authorizedpersonnel) is generally considered to besatisfactory. A 100 % solution would be toinstall overall protection from the HV circuitbreaker to the LV circuit breaker, as providedby the differential protective schemepreviously described.

HV earth-fault relay settingsEarth-fault relays have low current-settingranges, i.e. they are sensitive instruments,and can, in consequence, act to clear adeveloping short-circuit fault in its earlystages, thereby minimizing the damage toinsulation at the fault position, and reducingthe risk of fire.Care must be taken however, to avoidincreasing the sensitivity (by reducing theoperating-current setting) to a point wherethe relay can be caused to operate when anearth fault occurs on a nearby circuit, thecircuit on which the relay is installed being(perfectly) healthy (unfaulted).This false operation is due to the inherentcapacitance to earth of the power-systemphase conductors and connected loads, andis especially likely to occur on impedance-earthed systems (which are common at theHV levels covered by this guide).In normal circumstances, the capacitivecurrent from each phase to earth haspractically the same magnitude (Ic) and thethree currents summate in the earth to givethe so-called “residual” current, which,because of the balanced conditions (in thiscase) is theoretically zero.A short-circuit of one phase to earth, asalready mentioned in “Earth faults on IT-earthed systems” (Sub-clause 3.1) will cause:c the voltage of the faulted conductor, andthat of all conductors of the faulty phase overa wide area surrounding the fault location, tofall to practically zero volts,c the voltage of the two healthy phaseconductors over the same area to increase by(up to) etimes their original value withrespect to earth.As a result of this, the residual current in allcircuits affected by the voltage changes willno longer approximate to zero. On healthycircuits close to the fault position, the residualcurrent will, in fact, amount to almost 3 Ic, asillustrated in figure C22.If a healthy circuit has a significantcapacitance to earth (a long overhead line ora section of underground cable) then 3 Ic willbe detected by its earth-fault relay which, ifgiven an oversensitive setting, will trip thecircuit breaker of the unfaulted circuit.A conventional minimum value of earth-fault-current setting intended to avoid this problem,commonly adopted in the power-distributionindustry, is 6 Ic (i.e. a safety factor of 2).This phenomenon is only of interest to a HVinstallation-design engineer in cases wherethe HV circuit breaker and protective relays

are some distance from the transformer,especially if the supply is by undergroundcable, and the HV nominal voltage is high,e.g. u 20 kV.The foregoing discussion of the presence ofresidual capacitive components in the earth-fault current of impedance-earthed systems,leads conveniently to an explanation of theprinciple of the Petersen coil.

System earthing on overhead-line HVsystems by means of a Petersen coilIn the resistance-earthed system describedabove, it was shown that the current throughthe earth fault is the sum of the residualcapacitive currents of the system and thecurrent which flows through the resistor.This resistor current is in phase with thevoltage of the faulted phase (the voltagevector being reversed during the fault period)as shown in the vector diagram of figureC22. The phase displacement between theresistor current and the residual capacitivecurrent is practically 90 degrees.If now, the resistor is replaced by a reactor,the reactor current will lag the (reversed)faulted phase voltage by 90 degrees and will,therefore, be in phase opposition to theresidual capacitive current.By a suitable choice of reactance value theresidual capacitive current through the faultcan, in principle, be exactly cancelled, i.e.there will be no fault current flowing to earth,as shown in figure C23. This is the principleof Petersen Coil operation.In practice, absolute cancellation of faultcurrent is not possible, owing to conductor-and fault- path resistances, and to theimpracticality of precise coil tuning at alltimes. These effects prevent perfect mutualcancellation of the opposing currents, butproviding the fault-path resistance is low, thefault current can be reduced to very smallvalues.Petersen coils are provided with a number oftapping steps on the coil to cover a range ofsystem capacitance values. These coils areassociated mainly with isolated O/H linepower networks in the HV voltage rangescovered by this guide.


COperational advantagesAdvantages of the system include:c continuity of supply in the (common) eventof an earth fault. In principle, the system canoperate indefinitely with an earth fault onone phase,

fig. C22: earth-fault diagram.

3

2

1

3

2

1

3

2

1

eIC2A

eIC1A

0 V3 0

eIC2B

eIC1B

0 V3 0IF

IF

eIC2C

eIC1C

0 V3 0

IFI’F

powersupplysource

earth-fault protection relays of the healthy circuits A and C will detect an apparent earth fault

3ICA

3ICB

IF

3ICC

voltages during short-circuit to earth on phase 3

V2V3 = 0

V1

VNE

I’F

I’F

IF

IF

eIc2

eIc1power supply

simplified diagram showing current division

normal voltages and capacitive currents

V2V3

V1

IC2

IC1

IC3

residual current on healthy circuits during fault

V2V3 = 0

V13IC

√3IC1

√3IC2

V2

V13IC

√3IC1

√3IC2

VNE

I'F

IF

fault current IF is the vector sum of the neutral resistor current I’F and the residual capacitive currents of the system 3IC

A

B

C

eIC1 + eIC2 = 3ICI’F >> 3ICIF = I’F + 3IC

c damage at the fault position is limited,owing to the restricted current level,c disturbance to neighbouring systems atthe instant of fault is practically non-existent.


C



I’F

I’F = eIC1 + eIC2 = 3IC

IF = 0

eIc2

eIc1power supply

simplified diagram showing current division when XC / XL = 3 (where XC = capacitive reactance of one phase to earth)

I’F = 3IC

LPetersen coil

0 0

voltages during short-circuit to earth on phase 3

V2V3 = 0

V1

VNE

normal voltages and capacitive currents

V2V3

V1

IC2

IC1

IC3

vector diagram for condition I’F = 3 IC

I’F

V2E

V13IC

eIC1

eIC2

VNE

residual current on healthy circuits during fault

V2V3 = 0

V13IC

√3IC1

eIC2

fig. C23: earth fault diagram (with Petersen coil).


C

The risk and consequences of a fire areparticularly serious.The prefabricated equipments are conceivedand manufactured in a way that avoidsexcessive temperature rise in normal use.Where an installation includes one or severalliquid-insulated transformers, the regulationsand arrangements relative to the protectionand construction details must be fullyrespected, and are described in Sub-clause 4.3: “choice of HV/LV transformers”.

3.3 protection against thermal effects

3.4 interlocks and conditioned manœuvres

Mechanical and electrical interlocks areincluded on mechanisms and in the controlcircuits of apparatus installed in substations,as a measure of protection against anincorrect sequence of manœuvres byoperating personnel.Mechanical protection is afforded by:c compartments enclosing specific parts ofequipment in pre-fabricated HV cells,c key-transfer interlocking.

key interlockingThe most widely-used form of locking/interlocking depends on the principle of key-transfer.The principle is based on the possibility offreeing or trapping one or several keys,according to whether or not the conditions ofsafety are satisfied.These conditions can be combined in uniqueand obligatory sequences, therebyguaranteeing the safety of personnel by theavoidance of an incorrect operationalprocedure.For example, access to a HV panel requiresa certain number of operations which must becarried out in a pre-determined order. It isnecessary to carry out manœuvres in thereverse order to restore the system to itsformer condition.Non-observance of the correct sequence ofmanœuvres in either case may haveextremely serious consequences for theoperating personnel, as well as for theequipment concerned.

Note: It is important to provide for a schemeof interlocking in the basic design stage ofplanning a HV/LV substation.In this way, the apparatuses concerned willbe equipped during manufacture in acoherent manner, with assured compatibilityof keys and locking devices.

interlocks in substations

equipped with metalclad

switchgear

In a HV/LV distribution substation whichincludes:c a single incoming HV panel or two incomingpanels (from parallel feeders) or twoincoming/outgoing ring-main panels,c a transformer switchgear-and-protectionpanel, which can include a load-break/isolating switch with HV fuses and anearthing switch, or a circuit breaker and line-isolating switch together with an earthingswitch,

an interlocking scheme is intended toprevent any operational manœuvrewhich would expose operatingpersonnel to danger.

c a transformer compartment interlocks allowmanœuvres and access to different panels inthe following conditions:v operation of the load-break/isolatingswitchif the panel door is closed and the associatedearthing switch is open,v operation of the line-isolating switch ofthe transformer switchgear - and -protection panelif the door of the panel is closed, and- if the circuit breaker is open, and theearthing switch(es) is (are) open.v closure of an earthing switchif the associated isolating switch(es) is (are)open*,v access to the interior of each panelif the isolating switch for the panel is openand the earthing switch(es) in the panel is(are) closed,v closure of the door of each panel orcompartmentif the earthing switch(es) in the panel is (are)closed,v access to the HV fuses of a substationsupplied by two incomers from parallelfeedersif the two isolating switches are open and thetwo earthing switches in the panel are closed,v access to the compartment(s) occupiedby the VT(s)if the HV isolating switch is open, and if theLV isolating device is open,v operation of the isolating switches in theVT panelif the door of the panel is closed.* if the earthing switch is on an incoming circuit, theassociated isolating switches are those at both ends of thecircuit, and these should be suitably interlocked.


C


3.4 interlocks and conditioned manœuvres (continued)

practical example

In a consumer-type substation with LVmetering, the interlocking scheme mostcommonly used is HV/LV/TR (high voltage/low voltage/transformer).The aim of the interlocking is:c to prevent access to the transformercompartment if the earthing switch has notbeen previously closed,c to prevent the closure of the earthing switchin a transformer switchgear-and-protectionpanel, if the LV circuit breaker of thetransformer has not been previously locked“open” or “withdrawn”.Access to the HV or LV terminals of atransformer, protected upstream by a HVswitchgear-and-protection panel, containing aHV load-break / isolating switch, HV fuses,and a HV earthing switch) must comply withthe strict procedure described below, and isillustrated by the diagrams of figure C24.Note: The transformer in this example isprovided with plug-on type HV terminalconnectors which can only be removed byunlocking a retaining device common to allthree phase connectors*.The HV load-break / isolating switch ismechanically linked with the HV earthingswitch such that only one of the switches canbe closed, i.e. closure of one switchautomatically blocks the closure of the other.* or may be provided with a common protective cover overthe three terminals.

Procedure for the isolation and earthing ofthe power transformer, and removal of theHV plug-type shrouded terminalconnections (or protective cover).c initial conditions:v HV load-break/isolating switch and LVcircuit breaker are closed,v HV earthing switch locked in the openposition by key “O”,v key “O” is trapped in the LV circuit breakeras long as that circuit breaker is closed,c step 1:v open LV CB and lock it open with key “O”,v key “O” is then released,c step 2:v open the HV switch,v check that the “voltage presence” neonindicators extinguish when the HV switch isopened,c step 3:v unlock the HV earthing switch with key “O”and close the earthing switch,v key “O” is now trapped,c step 4:v the access panel to the HV fuses can nowbe removed (i.e. is released by closure of theHV earthing switch). Key “S” is located in thispanel, and is trapped when the HV switch isclosed,v turn key “S” to lock the HV switch in theopen position,v key “S” is now released,c step 5:v key “S” allows removal of the commonlocking device of the plug-type HV terminalconnectors on the transformer or of thecommon protective cover over the terminals,as the case may be.In either case, exposure of one or moreterminals will trap key “S” in the interlock.

fig. C24: example of HV/LV/TRinterlocking.

S

S

S

S

S

S

panel or door

legend

HV switch and LV CB closed

HV fuses accessible

transformer HV terminals accessible

key absentkey freekey trapped

O

O

O

O


CThe result of the foregoing procedure is that:a) the HV switch is locked in the openposition by key “S”.Key “S” is trapped at the transformerterminals interlock as long as the terminalsare exposed.b) the HV earthing switch is in the closedposition but not locked, i.e. may be opened orclosed. When carrying out maintenance work,a padlock is generally used to lock theearthing switch in the closed position, the keyof the padlock being held by the engineersupervizing the work.c) the LV CB is locked open by key “O”, whichis trapped by the closed HV earthing switch.

The transformer is therefore safely isolatedand earthed.It may be noted that the upstream terminal ofthe load-break switch may remain alive in theprocedure described. There are threereasons for this:c the terminals in question are located in aseparate inaccessible compartment in theparticular switchgear under discussion,c the open contacts of the switch have anearthed screen interposed between them,c the envelope containing the switch ismoulded from insulating material, filled withSF6 gas, and sealed for life.In the general case, the upstream terminalsof such a switch (or circuit breaker) will beexposed within the compartment, and anearthing switch will be provided, interlockedmechanically with a line-isolating switch. Oragain, it may be necessary (depending on thetype of switchgear) to isolate and lock off theincoming supply cable at its remote end,before closing the local earthing switch.Any scheme of interlocking must evidentlyinclude appropriate procedures, along similarlines to those described above.


C

4. the consumer substation with LV metering

A consumer substation with LV metering is anelectrical installation connected to a public-supply system at a nominal voltageof 1 kV - 35 kV, and includes a single HV/LVtransformer generally not exceeding1,250 kVA.

4.1 general

functions

The substationAll component parts of the substation arelocated in one chamber, either in an existingbuilding, or in the form of a prefabricatedhousing exterior to the building.Connection to the HV networkConnection at HV can be:c either by a single service cable or over-head line, or,c via two mechanically interlocked load-breakswitches with two service cables fromduplicate supply feeders, or,c via two load-break switches of a ring-mainunit.

The transformerSince the use of PCB*-filled transformers isprohibited in most countries, the preferredavailable technologies are:c oil-immersed transformers for substationslocated outside premises,c dry-type, vacuum-cast-resin transformersfor locations inside premises, e.g. multi-storeyed buildings, buildings receiving thepublic, and so on...

* polychlorinated biphenyl.

MeteringMetering at low voltage allows the use ofsmall metering transformers at modest cost.Most tariff structures take account oftransformer losses.

LV installation circuitsA low-voltage circuit breaker, suitable forisolation duty with visible contacts and lockingoff facilities, to:c supply a distribution board,c protect the transformer against overloadingand the downstream circuits against short-circuit faults.

one-line diagrams

The diagrams on the following page(figure C25) represent:c the different methods of HV serviceconnection, which may be one of four types:v single-circuit service,v single-circuit (for later change to ring-mainservice),v duplicate service (interlockedmechanically),v ring-main service,c HV protective functions and HV/LVtransformation,c LV metering and general isolation functions,c LV protective and distribution functions,c zones of access for interested parties.


C

power supply system service connection HV protection andHV/LV transformation

LV meteringand isolation

LV distribution and protection

supplier/consumer interface transformer LV terminals downstream terminals of LV isolator

protection

(permitted if IHV nominali 45 A and one transformer )

protection

protection

duplicate-supply service

single-line service (equipped for extension to form a ring main)

single-line service

protection

protection+

auto-changeover switch

automatic LVstandby source

consumer

power-supply authority

authorized access limits

testing authority

ring-main service

consumer

(permitted if IHV nominali 45 A and one transformer )

(always permitted)

fig. C25: consumer substation with LV metering.


C

4. the consumer substation with LV metering (continued)

standards and specifications

The SF6 switchgear and equipmentsdescribed below are rated for 1 kV - 24 kVsystems and conform to the followinginternational and national standards:c international:IEC 56-1, 129, 265-1, 298, 694,c national: French: UTE, EDF,

British: BS,German: VDE,American: ANSI.

4.2 choice of panels

type of material

All kinds of switchgear arrangements arepossible when using modular compartmentedpanels, and provisions for later extensionsare easily realized.Compact substations of modular panels areparticularly applicable in the following cases:c ring-main substations (a 3-functionmonobloc assembly),c severe climatic or heavily-pollutedconditions (integral insulation),c insufficient space for “classical”switchboards.This “all - SF6” equipment is distinguished byits reduced dimensions, its integratedfunctions and by its operational flexibility.

operational safety of

compartmented metalclad panels

DescriptionThe following notes describe a “state-of-the-art” load-break / isolating-switch panel (seefigure C26) incorporating the most moderndevelopments for ensuring:c operational safety,c minimum space requirements,c extendibility and flexibility,c minimum maintenance requirements.Each panel includes 4 compartments:c switchgear: the load-break switch isincorporated in an SF6-filled hermeticallysealed (for life) molded epoxy-resin unit,c connections: by cable at terminals locatedon the molded load-break switch unit,c busbars: modular, such that any number ofpanels may be assembled side-by-side toform a continuous switchboard,c control and indication: a control andinstrument compartment which canaccommodate automatic control and relayingequipment. An additional compartment maybe mounted above the existing one ifrequired.

Cable connectionsCable connections are provided inside acable-terminating compartment at the front ofthe unit, to which access is gained byremoval of the front panel of thecompartment.The units are connected electrically bymeans of prefabricated sections of busbars.Site erection is effected by following theassembly instructions.Operation of the switchgear is simplified bythe grouping of all controls and indications ona control panel at the front of each unit.The technology of these switchgear units isessentially based on operational safety, easeof installation and low maintenancerequirements.

fig. C26: compartmented SF6 HVload-break isolating switch.


Cstate of isolation clearly apparent

The load-break/isolating switch fully satisfiesthe requirement of “isolation clearly apparent"as defined in IEC 129, by means of:c a position indicator accurately reflecting theopen state of the contacts,c an earthed metal barrier interposedbetween the open contacts.Interlocksc closure of the switch is not possible unlessthe earth switch is open and the access panelto the cable-terminations compartment* isclosed,c closure of the earthing switch is onlypossible if the load-break/isolating switch isopen,c opening of the access panel to the cable-terminations compartment* is only possible ifthe earthing switch is closed,c the load-break/isolating switch is blocked inthe open position when the above-mentionedaccess panel is open. Operation of theearthing switch is then possible.* where HV fuses are used they are located in thiscompartment.

Apart from the functional interlocks notedabove, each switchgear panel includes:c built-in padlocking facilities,c 5 predrilled sets of fixing holes for possiblefuture interlocking locks.

Manœuvresc operating handles, levers, etc. required forswitching manœuvres are grouped togetheron a clearly illustrated panel,c all closing-operation levers are identical onall units (except those containing a circuitbreaker),c operation of a closing lever requires verylittle effort,c opening or closing of a load-break/isolatingswitch can be by lever or by push-button forautomatic switches,c conditions of switches (Open, Closed,Spring-charged), are clearly indicated.

table C27: standard short-circuit MVA and current ratings at different levels of nominalvoltage.

(1) ITH/1 sec: thermal withstand current for 1 second

(2) Isc: short-circuit current(3) I

CL: peak rated closing current

Choice of short-circuit withstand ratings

short-circuit (MVA) for nominal system voltages I TH/1 sec. (1) ICL(3)

(kV) Isc (2) (kA)3 3.3 4.16 5 5.5 6 6.6 10 11 13.8 15 20 22 33 (kA) r.m.s. peak65 70 90 110 120 130 145 215 240 300 325 435 475 715 12.5 31.575 85 105 125 135 150 165 250 275 345 375 500 550 825 14.4 36.585 90 115 140 150 165 185 280 305 385 415 555 610 915 16 40110 120 150 180 200 220 240 365 400 500 545 20 50135 150 190 230 250 275 300 455 500 25 62.5165 180 227 275 300 330 360 31.5 79


C


Three types of HV switchgear panel aregenerally available:c load-break switch and separate HV fuses inthe panel,c load-break switch/HV fuses combination,c circuit breaker.Seven parameters influence the optimumchoice:c the primary current of the transformer,c the insulating medium of the transformer,c the position of the substation with respect

to the load centre,c the kVA rating of the transformer,c the distance from switchgear to thetransformer,c the use of separate protection relays (asopposed to direct-acting trip coils).

Note: the fuses used in the load-break/fuse-switch combination have striker-pins whichensure tripping of the 3-pole switch on theoperation of one (or more) fuse(s).

4.3 choice of HV switchgear panel for a transformer circuit

4.4 choice of HV/LV transformer

characteristic parameters

of a transformer

A transformer is characterized in part by itselectrical parameters, but also by itstechnology and its conditions of use.

Electrical characteristicsc rated power (Pn): the conventionalapparent-power in kVA on which otherdesign-parameter values and theconstruction of the transformer are based.Manufacturing tests and guarantees arereferred to this rating,c frequency: for power distribution systems ofthe kind discussed in this guide, thefrequency will be 50 Hz or 60 Hz,c rated primary and secondary voltages: for aprimary winding capable of operating at morethan one voltage level, a kVA ratingcorresponding to each level must be given.The secondary rated voltage is its open-circuit value,c rated insulation levels: are given byovervoltage-withstand test values at powerfrequency, and by high-voltage impulse testswhich simulate lightning discharges. At thevoltage levels discussed in this guide,overvoltages caused by HV switchingoperations are generally less severe thanthose due to lightning, so that no separatetests for switching-surge withstand capabilityare made.IEC standards define the rated (power-frequency) voltage and the “highest voltagefor equipment” in exactly the same terms, asnoted in Sub-clause 1.1 of this Chapter,c off-circuit tap-selector switch: generallyallows a choice of up to ± 2.5 % and ± 5 %level about the rated voltage of the highestvoltage winding. The transformer must bede-energized before this switch is operated;however, on-load tap-changers are available(e.g. ± 12.5%) where circumstances requireit,c winding configurations: are indicated indiagrammatic form by standard symbols forstar, delta and inter-connected-star windings;(and combinations of these for special duty,e.g. six-or twelve-phase rectifier transformers,etc.) and in an IEC-recommended alpha-numeric code. This code is read from left-to-right, the first letter refers to the highestvoltage winding, the second letter to the nexthighest, and so on,

v capital letters refer to the highest voltagewindingD = deltaY = starZ = interconnected-star (or zigzag)N = neutral connection brought out to a terminalv lower-case letters are used for tertiary andsecondary windingsd = deltay = starz = interconnected-star (or zigzag)n = neutral connection brought out to a terminalv a number from 0 to 11, corresponding tothose, on a clock dial (“O” is used instead of“12”) follows any pair of letters to indicate thephase change (if any) which occurs duringthe transformation. If a neutral terminal isavailable then the number appears after theN (or n).A very common winding configuration usedfor distribution transformers is that of aDyn 11 transformer, which has a delta HVwinding with a star-connected secondarywinding the neutral point of which is broughtout to a terminal. The phase change throughthe transformer is +30 degrees, i.e. phase 1secondary voltage is at “11 o’clock” whenphase 1 of the primary voltage is at“12 o’clock”, as shown in figure C36. Allcombinations of delta, star and zigzagwindings produce a phase change which (ifnot zero) is either 30 degrees or a multiple of30 degrees.IEC 76-4 describes the “clock code” in detail.

Characteristics related to the technologyand utilization of the transformerThis list is not exhaustive:c choice of technology.The insulating medium is:v liquid (mineral oil) or,v solid (epoxy resin and air),c for interior or exterior installation,c altitude (i 1,000 m is standard),c temperature (IEC 76-2),v maximum ambient air: 40 °C,v daily maximum average ambient air: 30 °C,v annual maximum average ambient air:20 °C.For non-standard operating conditions, referto C1.1 "Influence of the Ambienttemperature and altitude on the ratedcurrent".

a transformer is characterized in partby its electrical parameters, but alsoby its technology and its conditions ofuse.

the rated power of the transformer ischosen according to the maximumapparent power, as determined inB.4.6.


Cdescription of insulation

techniques

There are two basic classes of distributiontransformer presently available:c dry type (cast in resin),c liquid filled (oil-immersed).Dry type transformersThe windings of these transformers areinsulated by resin cast under vacuum (whichis patented by major manufacturers).It is recommended that the transformer bechosen according to the CENELECstandards documents HD 46451, as follows:c environment class E2 (frequentcondensation and/or high level of pollution),c climatic conditions class C2 (utilization,transport and stockage down to -25 °C),c fire resistance (transformers exposed to firerisk with low flammability and self-extinguishing in a given time).The following description refers to theprocess developed by a leading Europeanmanufacturer in this field.The encapsulation of a winding uses threecomponents:c epoxy-resin based on biphenol A with aviscosity that ensures complete impregnationof the windings,c anhydride hardener modified to introduce adegree of resilience in the moulding, essentialto avoid the development of cracks during thetemperature cycles occurring in normaloperation,c pulverulent additive composed oftrihydrated alumina Al (OH)3 and silica whichenhances its mechanical and thermalproperties, as well as giving exceptionalintrinsic qualities to the insulation in thepresence of heat.This three-component system ofencapsulation gives Class F insulation(∆θ = 100 K) with excellent fire-resistingqualities and immediate self-extinction. Thesetransformers are therefore classified as non-flammable.The mouldings of the windings contain nohalogen compounds (chlorine, bromine, etc.)or other compounds capable of producingcorrosive or toxic pollutants, therebyguaranteeing a high degree of safety topersonnel in emergency situations, notablyin the event of a fire.It also performs exceptionally well in hostileindustrial atmospheres of dust, humidity, etc.See figure C28.

Liquid-filled transformersThe most common insulating/cooling liquidused in transformers is mineral oil. Mineraloils are specified in IEC 296. Beingflammable, safety measures are obligatory inmany countries, especially for indoorsubstations. The DGPT unit (Detection ofGas, Pressure and Temperature) ensures theprotection of oil-filled transformers.In the event of an anomaly, the DGPT causesthe HV supply to the transformer to be cut offvery rapidly, before the situation becomesdangerous.Mineral oil is bio-degradable and does notcontain PCB (polychlorinated biphenyl),which was the reason for banning askerel,i.e. Pyralène, Pyrolio, Pyroline...

fig. C28: dry-type transformer.

On request, mineral oil can be replacedby an alternative insulating liquid, by adaptingthe transformer, as required, and takingappropriate additional precautionsif necessary.The insulating fluid also acts as a coolingmedium; it expands as the load and/or theambient temperature increases, so that allliquid-filled transformers must be designedto accommodate the extra volume of liquidwithout the pressure in the tank becomingexcessive.


C


4.4 choice of HV/LV transformer (continued)

There are two ways in which this pressurelimitation is commonly achieved:c hermetically-sealed totally-filled tank(up to 10 MVA at the present time)Developed by a leading French manufacturerin 1963, this method was adopted by thenational power authority in 1972, and is nowin world-wide service.Expansion of the liquid is compensated bythe elastic deformation of the oil-coolingpassages attached to the tank.The “total-fill” technique has many importantadvantages over other methods:v oxydation of the dielectric liquid (withatmospheric oxygen) is entirely precluded,v no need for an air-drying device, and so noconsequent maintenance (inspectionand changing of saturated dessicant),v no need for dielectric-strength testof the liquid for at least 10 years,v simplified protection against internal faultsby means of a DGPT device is possible,v simplicity of installation: lighter and lowerprofile (than tanks with a conservator) andaccess to the HV and LV terminals isunobstructed,v immediate detection of (even small) oilleaks; water cannot enter the tank.c air-breathing conservator-type tank atatmospheric pressureExpansion of the insulating liquid is taken upby a change in the level of liquid in anexpansion (conservator) tank, mountedabove the transformer main tank, as shown isfigure C30. The space above the liquid in theconservator may be filled with air which isdrawn in when the level of liquid falls, and ispartially expelled when the level rises. Whenthe air is drawn in from the surroundingatmosphere it is admitted through an oil seal,before passing through a dessicating device(generally containing silica-gel crystals)before entering the conservator. In somedesigns of larger transformers the spaceabove the oil is occupied by an impermeableair bag so that the insulation liquid is never incontact with the atmosphere. The air entersand exits from the deformable bag through anoil seal and dessicator, as previouslydescribed. A conservator expansion tank isobligatory for transformers rated above10 MVA (which is presently the upper limitfor “total-fill” type transformers).

fig. C30: air-breathing conservator-typetank at atmosphere pressure.

fig. C29: hermetically-sealed totally-filledtank.


Cchoice of technology

As discussed above, the choice oftransformer is between liquid-filled or dry-type. For ratings up to 10 MVA, totally-filledunits are available as an alternativeto conservator-type transformers.A choice depends on a numberof considerations, including:c safety of persons in proximity to thetransformer. Local regulations and officialrecommendations may have to be respected,c economic considerations, taking accountof the relative advantages of each technique.Regulations affecting the choicec dry-type transformer:v in some countries a dry-type transformer isobligatory in high apartment blocks,v dry-type transformers impose noconstraints in other situations,c transformers with liquid insulation:v this type of transformer is generallyforbidden in high apartment blocks,v for different kinds of insulation liquids,installation restrictions, or minimum protect-against fire risk, vary according to the classof insulation used,c some countries in which the use of liquiddielectrics is highly developed, classify theseveral categories of liquid according to theirfire performance. This latter is assessedaccording to two criteria: the flash-pointtemperature, and the minimum calorificpower. The principal categories are shown inTable C31 in which a classification code isused for convenience.

National standards exist which define theconditions for the installation of liquid-filledtransformers. No equivalent IEC standardhas yet been established.The national standard is aimed at ensuringthe safety of persons and property andrecommends, notably, the minimummeasures to be taken against the risk of fire.The main precautions to observeare indicated in Table C32.c for liquid dielectrics of class L3 thereare no special measures to be taken,c for dielectrics of classes 01 and K1 themeasures indicated are applicable only ifthere are more than 25 litres of dielectricliquid in the transformer,c for dielectrics of classes K2 and K3 themeasures indicated are applicable onlyif there are more than 50 litres of dielectricliquid in the transformer.

table C31: categories of dielectric fluids.

code dielectric fluid flash-point ( °C) minimum calorific power (MJ/kg)O1 mineral oil < 300 -K1 high-density hydrocarbons > 300 48K2 esters > 300 34 - 37K3 silicones > 300 27 - 28L3 insulating halogen liquids - 12


C


4.4 choice of HV/LV transformer (continued)

table C32: safety measures recommended in electrical installations using dielectric liquids of classes 01, K1, K2 or K3.

Measure 1: arrangements such that if thedielectric escapes from the transformer, it willbe completely contained (in a sump, by sillsaround the transformer, and by blocking ofcable trenches, ducts and so on, duringconstruction).Measure 1A: in addition to measure 1,arrange that, in the event of liquid ignitionthere is no possibility of the fire spreading(any combustible material must be movedto a distance of at least 4 metres fromthe transformer, or at least 2 metres fromit if a fire-proof screen [of 1 hour rating] isinterposed).Measure 2: arrange that burning liquid willextinguish rapidly and naturally (by providinga pebble bed in the containment sump).Measure 3: an automatic device (DGPTor Buchholz) for cutting off the primary powersupply, and giving an alarm, if gas appears inthe transformer tank.Measure 4: automatic fire-detection devicesin close proximity to the transformer, forcutting off primary power supply, and givingan alarm.Measure 5: automatic closure by fire-proofpanels (1/2 hour minimum rating) of allopenings (ventilation louvres, etc.) in thewalls and ceiling of the substation chamber.

Notes:(a) a fire-proof door (rated at 2 hours)is not considered to be an opening.(b) - transformer chamber adjoining aworkshop and separated from it by walls,the fire-proof characteristics of which arenot rated for 2 hours,- areas situated in the middle of workshopsthe material being placed (or not) in aprotective container.(c) it is indispensable that the equipment beenclosed in a chamber, the walls of which aresolid, the only orifices being those necessaryfor ventilation purposes.

class no. of locationsof litres chamber or enclosed area reserved to qualified reserved to trained personnel other chambersdielec- above and authorized personnel, and separated from any and isolated from work areas or locationstric which other building by a distance D by fire-proof walls (2 hours rating) (b)fluid measures D > 8 m 4 m < D < 8 m D < 4 m (a) in the direc- no openings with opening(s)

must be tion of occupied areastaken

O1 25 no special interposition of fire-proof wall measures (1 + 2) measures (1 + 2 + 5) measuresK1 measures a fire-proof screen (2 hour rating) or 3 or 3 (1A + 2 + 4)(c)

(1 hour rating) against adjoining or 4 or (4 + 5) or 3building

K2 50 no special measures interposition of a no special measures 1A measures 1K3 fire-proof screen measures or 3 or 3

(1 hour rating) or 4 or 4L3 no special measures


Cthe determination

of optimal power

Oversizing a transformer results in:c excessive investment and unecessarilyhigh no-load losses, but,c lower on-load losses.Undersizing a transformer, causes:c a reduced efficiency when fully loaded,(the highest efficiency is attained in the range50% - 70% full load) so that the optimumloading is not achieved,c on long-term overload, seriousconsequences for:v the transformer, owing to the prematureageing of the windings insulation, and inextreme cases, resulting in failure ofinsulation and loss of the transformer,v the installation, if overheating of thetransformer causes protective relays to tripthe controlling circuit breaker.

Definition of optimal powerIn order to select an optimal power (kVA)rating for a transformer, the following factorsmust be taken into account:c list the power of installed power-consumingequipment as described in Chapter B,c decide the utilization (or demand) factor foreach individual item of load,c determine the load cycle of the installation,noting the duration of loads and overloads,c arrange for power-factor correction,if justified, in order to:v reduce cost penalties in tariffs based,in part, on maximum kVA demand,v reduce the value of declared load(P(kVA) = P (kW)/cos ø),c select, among the range of standardtransformer ratings available, taking intoaccount all possible future extensionsto the installation.It is important to ensure that coolingarrangements for the transformerare adequate.

the chamber is ventilated by naturalconvection or forced ventilation.

ventilation orifices

In the general case of cooling by natural aircirculation (AN) the ventilation of the chamberis arranged to remove the heat (producedby losses in the transformer) by naturalconvection.A good system of ventilation allows cool air toenter through an orifice of sectional area S atfloor level, and to leave the chamber throughan orifice of sectional area S' on the oppositewall to that of the air entry and at a height Habove the incoming-air orifice, as shown infigure C33.It is important to note that any restriction tothe free flow of a sufficient volume of air willresult in a reduction of power available fromthe transformer, if the rated temperature limitis not to be exceeded.

Natural ventilationThe formulae for calculating the sectionalarea of the ventilation orifices are as follows:S = 0.18 P/√H and S’ = 1.1 SWhere:P = the sum of the no-load losses and thefull-load losses expressed in kWS = the sectional area of the incoming-airorifice (area of louvres or grill to be deducted)expressed in mm2

S’ = the sectional area of the outgoing-airorifice (area of louvres or grill to be deducted)expressed in mm2

H = height (centre to centre) of the outgoingair orifice above the incoming-air orifice,expressed in metres.The formulae are valid for a mean ambienttemperature of 20 °C and up to an altitudeof 1,000 m.

Forced ventilationForced (i.e. electric-fan assisted) ventilationof the chamber is necessary for ambienttemperatures exceeding 20 °C, or if thechamber is badly ventilated; frequentoverloading of the transformer, and so on...The fan can be controlled by thermostat.Recommended air-flow rate, in cubic metresper second at 20 °C:c totally-filled transformer: 0.081 P,c dry-type Class F transformer: 0.05 Pwhere P = total losses in kW.

fig. C33: natural ventilation.

S'

S

H


C

5. a consumer substation with HV metering

A consumer substation with HV metering isan electrical installation connected to a publicsupply system at a nominal voltage of 1 kV -35 kV and generally includes a single HV/LVtransformer which exceeds 1,250 kVA, orseveral smaller transformers.The rated current of the HV switchgear doesnot normally exceed 400 A.

5.1 general

functions

The substationAccording to the complexity of the installationand the manner in which the load is divided,the substation:c might include one room containing the HVswitchboard and metering panel(s), togetherwith the transformer(s) and low-voltage maindistribution board(s),c or might supply one or more transformerrooms, which include local LV distributionboards, supplied at HV from switchgear in amain substation, similar to that describedabove.These substations may be installed, either:c inside a building, orc outdoors in prefabricated housings.Connection to the HV networkConnection at HV can be:c either by a single service cable or overheadline, orc via two mechanically interlocked load-breakswitches with two service cables fromduplicate supply feeders, orc via two load-break switches of a ring-mainunit.

MeteringBefore the installation project begins, theagreement of the power-supply authorityregarding metering arrangements must beobtained.A metering panel will be incorporated in theHV switchboard. Voltage transformers andcurrent transformers, having the necessarymetering accuracy, may be included in themain incoming circuit breaker panel or (in thecase of the voltage transformer) may beinstalled separately in the metering panel.

Transformer roomsIf the installation includes a number oftransformer rooms, HV supplies from themain substation may be by simple radialfeeders connected directly to thetransformers, or by duplicate feeders to eachroom, or again, by a ring-main, according tothe degree of supply security desired.In the two latter cases, 3-panel ring-mainunits will be required at each transformerroom.

Local emergency generatorsEmergency standby generators are intendedto maintain a power supply to essential loads,in the event of failure of the power supplysystem.

CapacitorsCapacitors will be installed, according torequirements:c in stepped HV banks at the mainsubstation, orc at LV in transformer rooms.TransformersFor additional supply-security reasons,transformers may be arranged for automaticchangeover operation, or for paralleloperation.

one-line diagramsThe diagrams shown in figure C34 represent:c the different methods of HV serviceconnection, which may be one of four types:v single-circuit service,v single-circuit (for later change to ring-mainservice),v duplicate service (interlockedmechanically),v ring-main service,c general protection at HV, and HV meteringfunctions,c protection of outgoing HV circuits,c protection of LV distribution circuits.


C

power supply system

service connection HV protection and metering

HV distribution and protection of outgoing circuits

LV distribution and protection

supplier/consumer interface downstream terminals of HV isolator for the installation

LV terminals of transformer

duplicate-supply service

single-line service(equipped for extension to form a ring main)

single-lineservice protection

LV

protection+

automaticchangeover

feature

automatic LV standby source

consumer

testing authority

power-supply authority

authorized access limits

ring-main service

protection

automatic LV/HV standby source

HV LV

I nominal of transformer u 45 A

a single transformer

fig. C34: consumer substation with HV metering.


C

5. a consumer substation with HV metering (continued)

A substation with HV metering includes, inaddition to the panels described in 4.2,panels specifically designed for metering and,if required, for automatic or manualchangeover from one source to another.

metering and general protection

These two functions are achieved by theassociation of two panels:c one panel containing the VT,c the main HV circuit breaker panel,containing the CTs for measurementand protection.The general protection is usually againstovercurrent (overload and short-circuit) andearth faults. Both schemes use protectiverelays which are sealed by the power-supplyauthority.

5.2 choice of panels

power-supply changeover

schemes

Some national standards recommend asupplementary protection when aninstallation includes an emergency automaticchangeover to a local generator. Thestandard states that operation of the standbyplant must not, in any circumstances, result inperturbations on the power-supply network.This means that, apart from protectivedevices intended to protect the generator:c either a scheme of interlocking mustpreclude any possibility of parallel operationof the generator with the power system, orc a suitable automatic de-coupling schemeagreed with the power-supply authority, whichwill trip the paralleling circuit breaker in theevent of a short-circuit, or other anomaly,occurring on the power supply system, or onthe installation.

In the second case, the tripping command tothe decoupling circuit breaker must operatereliably on undervoltage and reverse-powerprotection. Setting of protection relays iscarried out by the power-supply authorityand made inaccessible to the consumerby sealing, or some equivalent means.The tripping supply and switchgear-controlswitch(es) must also be inaccessible to theconsumer.

fig. C35: typical arrangement ofswitchgear panels for HV metering.

fig. C36: section of HV switchboard including standby supply panel.

metering VT panel

main HV CB panel with metering and protection CTs

HV distribution panels

power-supply network

from standby generator P i 20,000 kVA

HV distribution panels for which standby supply is required

automatic changeover panel

busbar transition panel

to remainder of the HV switchboard


Csmall generators operating

in parallel with public supply

networks

The following notes indicate some basicconsiderations to be taken into account whenparallel operation of consumer’s generatorswith the public power-supply networks isplanned.A voltage regulator controlling an alternator isgenerally arranged to respond to a reductionof voltage at its terminals by automaticallyincreasing the excitation current of thealternator, until the voltage is restored tonormal. When it is intended that the alternatorshould operate in parallel with others, theAVR (Automatic Voltage Regulator) isswitched to “parallel operation” in which theAVR control circuit is slightly modified(compounded) to ensure satisfactory sharingof kvars with the other parallel machines.When a number of alternators are operatingin parallel under AVR control, an increase inthe excitation current of one of them (forexample, carried out manually after switchingits AVR to Manual control) will havepractically no effect on the voltage level.In fact, the alternator in question will simplyoperate at a lower power factor (more kVA,and therefore more current) than before.The power factor of all the other machineswill automatically improve, such that the loadpower factor requirements are satisfied,as before.Consider now the case of a standbygenerator at a consumer’s substation, whichis operating in parallel with all the generationof the public power supply system.Supposing the power system voltage isreduced for operational reasons (it iscommon to operate HV systems within arange of ± 5% of nominal voltage, or evenmore, where load-flow patterns require it).An AVR set to maintain the voltage within± 3% (for example) will immediately attemptto raise the voltage by increasing theexcitation current of the alternator.Instead of raising the voltage, the alternatorwill simply operate at a lower power factorthan before, thereby increasing its currentoutput, and will continue to do so, until it iseventually tripped out by its overcurrentprotective relays.This is a well-known problem and is usuallyovercome by the provision of a “constant-power-factor” control switch on the AVR unit.By making this selection, the AVR willautomatically adjust the excitation current tomatch whatever voltage exists on the powersystem, while at the same time maintainingthe power factor of the alternator constant atthe pre-set value (selected on the AVRcontrol unit).In the event that the alternator becomes de-coupled from the power-system, the AVRmust be automatically (rapidly) switched backto “constant-voltage” control.

Note: The problem is essentially that of a“small” generator and a “large” system. Theterms “small” and “large” are relative, so that,for example, even a small generator at theend of a long line could probably operatesatisfactorily on constant-voltage control.To be more specific, if the system impedanceviewed from the generator location is high(i.e. the short-circuit fault level is low) thenconstant-voltage control may be satisfactory.In highly-developed networks where theshort-circuit fault levels are high, thenconstant-power-factor operation is generallyobligatory.A technical discussion with the power-supplyauthority will be necessary to resolve thequestion, but it is recommended that facilitiesfor both kinds of control be specified whenpurchasing the generator sets.


C

5. a consumer substation with HV metering (continued)

The need for operation of two or moretransformers in parallel often arises due to:c load growth, which exceeds the capactiyof an existing transformer,c lack of space (height) for one largetransformer,c a measure of security (the probabilityof two transformers failing at the same timeis very small),c the adoption of a standard sizeof transformer throughout an installation.

5.3 parallel operation of transformers

total power (kVA)

The total power (kVA) available when two ormore transformers of the same kVA rating areconnected in parallel, is equal to the sum ofthe individual ratings, providing that thepercentage impedances are all equaland the voltage ratios are identical.Transformers of unequal kVA ratings willshare a load practically (but not exactly)in proportion to their ratings, providingthat the voltage ratios are identical

and the per-centage impedances (at theirown kVA rating) are identical, or very nearlyso.In these cases, a total of more than 90% ofthe sum of the two ratings is normallyavailable.It is recommended that transformers, the kVAratings of which differ by more than 2: 1,should not be operated permanentlyin parallel.

conditions necessary for parallel

operation

All paralleled units must be supplied from thesame network.The inevitable circulating currents exchangedbetween the secondary circuits of paralleledtransformers will be negligibly small providingthat:c secondary cabling from the transformersto the point of paralleling have approximatelyequal lengths and characteristics,c the transformer manufacturer is fullyinformed of the duty intended for thetransformers, so that:

v the winding configurations (star, delta,zigzag star) of the several transformers havethe same phase change between primaryand secondary voltages,v the short-circuit per-centage impedancesare equal, or differ by less than 10%,v voltage differences between correspondingphases must not exceed 0.4%,v all possible information on the conditions ofuse, expected load cycles, etc. should begiven to the manufacturer with a view tooptimizing load and no-load losses.

common winding arrangements

As described in 4.4 “Electrical characteristics -winding configurations” the relationshipsbetween primary, secondary, and tertiarywindings depend on:c type of windings (delta, star, zigzag),c connection of the phase windings.Depending on which ends of the windingsform the star point (for example), a starwinding will produce voltages which are 180°displaced with respect to those produced ifthe opposite ends had been joined to formthe star point. Similar 180° changes occur inthe two possible ways of connecting phase-to-phase coils to form delta windings, whilefour different combinations of zigzagconnections are possible,c the phase displacement of the secondaryphase voltages with respect to thecorresponding primary phase voltages.As previously noted, this displacement (if notzero) will always be a multiple of 30° and willdepend on the two factors mentioned above,viz type of windings and connection(i.e. polarity) of the phase windings.By far the most common type of distributiontransformer winding configuration isthe Dyn 11 connection.

fig. C37: phase change through a Dyn 11transformer.

1

23

V12

1

2

3

N

1

2

3

voltage vectors

V12 on the primary winding produces V1N in the secondary winding and so on ...

windings correspondence

1

2

3

N


C

current transformers provided by power-supply authority

LV switchgear

LV connections from transformer

HV connections to transformer (included in a panel or free-standing)

2 incoming HV panels

HV switching and protection panel

connection to the power-supply network by single-core or three-core cables, with or without a cable trench

transformer LV cable trenchoil sump

6. constitution of HV/LV distribution substations

HV/LV substations are constructed accordingto the magnitude of the load and the kind ofpower system in question.Substations may be built in public places,such as parks, residential districts, etc. or onprivate premises, in which case the powersupply authority must have unrestrictedaccess. This is normally assured by locatingthe substation, such that one of its walls,which includes an access door, coincideswith the boundary of the consumers premisesand the public way.

6.1 different types of substation

Substations may be classified according tometering arrangements (HV or LV) and typeof supply (O/H ou U/G).The substations may be installed:c either indoors in chambers specially builtfor the purpose, or incorporated in anapartment block, etc., orc an outdoor installation mounted on a pole,or poles, (“H” structure or 4-polearrangement) or in a brick-built, concrete orprefabricated housing.Prefabricated housings mounted on aconcrete base provide a particularly simple,rapid and competitive choice.

6.2 indoor substations equipped with metal-enclosed switchgear

conception

Figure C38 shows a typical equipment layoutrecommended for a LV metering substation.

Remark: the use of a cast-resin dry-typetransformer would obviate the need for a fire-protection oil sump.

fig. C38: typical arrangment of switchgear panels for LV metering.


C

800 mini

meters

100

safety accessoriescommon earth busbar for the substation

6. constitution of HV/LV distribution substations (continued)

6.2 indoor substations equipped with metal-enclosed switchgear (continued)

service connections and

equipment interconnections

At high voltagec connections to the HV system are made by,and are the responsibility of the power-supplyauthority,c connections between the HV switchgearand the transformers may be:v by short copper bars where the transformeris housed in a panel forming part of the HVswitchboard,v by single-core unarmoured cables withsynthetic insulation,v single-core unarmoured cables to 250 A(or more) plug-in type terminals at thetransformer.At low voltagec connections between the LV terminals ofthe transformer and the LV switchgear maybe:v single-core unarmoured cables,v solid copper bars (circular or rectangularsection) with heat-shrinkable insulation.

Meteringc metering current transformers are generallyinstalled in the protective cover of the powertransformer LV terminals, the cover beingsealed by the supply authority,c alternatively, the current transformers areinstalled in a sealed compartment within themain LV distribution cabinet,c the meters are mounted on a panel which iscompletely free from vibrations,c placed as close to the current transformersas possible, andc are accessible only to the power-supplyauthority.The dials and graduations of the metersshould be at a height of approximately1.65 m. above floor level, not lower than0.7 m, and not higher than 1.8 m.

fig. C39: plan view of typical substation with LV metering.


Cearthing circuits

The substation must include:c an earth electrode for all exposedconductive parts of electrical equipment in thesubstation and exposed extraneous metalincluding:v protective metal screens,v reinforcing rods in the concrete base of thesubstation,v the common point of all current-transformersecondary windings.Note: Metal doors and ventilation louvres arenot connected to earth.c an earth electrode for the LV neutral pointof the transformer*,c removable links at strategic points formeasuring continuity and the resistances ofindividual electrodes,c an earth electrode for the installation*.* in small areas, the resistance zones of earth electrodesoverlap. In such cases all electrodes are interconnected toform a common earthing system for HV and LV equipments,as discussed in “Earthing connections” in Sub-clause 1.1 ofthis Chapter.

substation lighting

Supply to the lighting circuits can be takenfrom a point upstream or downstream of themain incoming LV circuit breaker. In eithercase, appropriate overcurrent protection mustbe provided. A separate automatic circuit (orcircuits) is (are) recommended for emergencylighting purposes.Operating switches, pushbuttons, etc. arenormally located immediately adjacent toentrances.Lighting fittings are arranged such that:c switchgear operating handles and position-indication markings are adequatelyilluminated,c all metering dials and instruction plaquesand so on, can be easily read.

materials for operation

and safety

The substation must be provided with:c materials for assuring safe exploitation ofthe equipment including:v a wooden stool and/or an insulating mat(rubber or synthetic),v a pair of insulated gloves stored in anenvelope provided for the purpose,v a voltage-detecting device for use on theHV equipment,v earthing attachments (according to type ofswitchgear),c fire-extinguishing devices of the powder orCO2 type,c warning signs, notices and safety alarms:v on the external face of all access doors, aDANGER warning plaque and prohibition ofentry notice, together with instructions forfirst-aid care for victims of electricalaccidents,v inside the substation: a first-aid panel asnoted above,v a DANGER plaque (skull and cross-bones,or a local equivalent sign) on each removablepanel providing access to live parts.


C


pole-top public distribution

substations

Field of applicationThese substations are mainly used to supplyisolated rural consumers from HV overheadline distribution systems:c at voltage levels between 1 - 24 kV,c from a single transformer not exceeding160 kVA and at a preferred LV voltage level of230/400 V (3-phase 4-wires),c with low-voltage metering.ConstitutionThese substations are commonly supplied bya single 3-wire line, with no local switchgearor fuses at the HV side of the transformer.Lightning arresters are provided, however, toprotect the transformer and consumers asshown in figure C40.Protection of the LV circuit is generallyprovided by two LV circuit breakers (D1) and(D2), shown in figure C41:c circuit breaker D1 protects the transformeragainst overloading and the LV serviceconnection against short-circuit faults. Thiscircuit breaker is mounted on the pole andhas inverse-time/current-relay trippingcharacteristics, or may be tripped by athermal-image relay monitoring thetransformer-windings temperature,c circuit breaker D2 is the main LV circuitbreaker for the installation.Tripping discrimination between these twocircuit breakers must be established, thesettings to be made by the power-supplyauthority, and sealed.

6.3 outdoor substations

General arrangement of equipmentAs previously noted the location of thesubstation must allow easy access, not onlyfor personnel but for equipment handling(raising the transformer, for example) and themanœuvring of heavy vehicles. Earthingelectrodes are commonly separated asdiscussed in Sub-clause 1.1 of this Chapter.See figure C42.

fig. C40: pole-mounted transformersubstation.

fig. C41: diagram showing the principles of a pole-mounted transformer substation.

fig. C42: separated earth electrodes.

Rp must have a maximum value derived in the same way asthat shown for Rs of case "E" in figure C7.RB < 3 ohms to limit the voltage appearing at the consumer'sinstallation in the event of a breakdown of HV/LV insulationdue to back-flashover, or other causes.

earthing conductor 25 mm2 copper

protective conductor cover

LV circuit breaker D1

safety earth mat

lightning arresters

remote pole-mounted load-break/isolating switch

pole-mounted transformer

LV circuit breaker at the transformer

metering

installation main circuit breaker

lightning arresters

D1 D2

installation

Rp RB RA


Cprefabricated housings

for outdoor substations

For more elaborate substations requiring theuse of ring-main units or a switchboard ofseveral circuit breakers, compactweatherproof and vermin-proof housings arecommonly used.These prefabricated units require theminimum civil work, being mounted on asimple concrete base, and are used for bothurban and rural substations.Among the advantages offered by theseunits, are:c an optimization of materials and safety by:v an appropriate choice from a wide range ofavailable housings,v conformity with all existing and foreseeableinternational standards,c a reduction in study and design time, and inthe cost of implementation, by:v minimal co-ordination between the severaldisciplines of building construction and siteworks,v realization, independent of the mainbuilding construction,v obviating the need for a temporary “hook-up” at the beginning of the site preparationwork,v simplification of civil work, which consistsonly of the provision of a reinforced-concreteplinth,c greatly simplified equipment installation andconnection.

fig. C43: cut-away view of typical HV/LV substation using a prefabricated housing.


C


6.3 outdoor substations (continued)

Other kinds of outdoor substation arecommon in some countries, based onweatherproof equipment exposed to theelements.These comprise a fenced area in which threeor more concrete plinths are installed:c for a ring-main unit, or one or more switch-fuse or circuit breaker unit(s),c for one or more transformer(s), andc for one or more LV distribution pillar(s).The simplicity of this arrangement iscountered by the high cost of weatherproofswitchgear, cable boxes, etc. and by anadverse visual impact.This class of substation is not favoured inresidential areas or in other locations wherevisual amenities are important, and for thesereasons have been largely supplanted byprefabricated housings and indoor-typeequipment in many countries.

low-voltage service connections

D

Low-voltage service connections

1. Low-voltage public distribution networks1.1 low-voltage consumers D11.2 LV distribution networks D71.3 the consumer-service connection D101.4 quality of supply voltage D13

2. Tariffs and metering D14

low-voltage service connections - D1

D

1. low-voltage public distribution networks

1.1 low-voltage consumers

the most-common LV supplies arewithin the range 120 V single phaseto 240/415 V 3-phase 4-wires.Loads up to 250 kVA can be suppliedat LV, but power-supply organizationsgenerally propose a HV service atload levels for which their LVnetworks are marginally adequate.An international voltage standard for3-phase 4-wire LV systems isrecommended by the IEC to be230/400 V.*

Low-voltage consumers are, by definition,those consumers whose loads can besatisfactorily supplied from the low-voltagesystem in their locality.The voltage of the local LV network may be120/208 V or 240/415 V, i.e. the lower orupper extremes of the most common 3-phaselevels in general use, or at some intermediatelevel, as shown in table D1.An international voltage standard for 3-phase4-wire LV systems is recommended by theIEC to be 230/400 V.*

Loads up to 250 kVA can be supplied at LV,but power-supply organizations generallypropose a HV service at load levels for whichtheir LV networks are marginally adequate.

country frequency domestic commercial industrial low-voltage& tolerance toleranceHz & % %

Australia 50 ± 0.1 240/415 (A) (E) 240/415 (A) 22 kV ± 6240 (L) 250/440 (A) 11 kV

440 (N) (6) 6.6 kV240/415 (A)250/440 (A)

Western 50 250/440 (A) (9) (9) ± 6 (10)Algeria 50 ± 1.5 127/220 (E) 220/380 (A) 10 kV + 5 and + 10

220 (L) (1) 127/220 (A) 5.5 kV6.6 kV220/380 (A)

Argentina 50 ± 1.0 225 (L) (1) 225/390 (A) 13.2 kV220 (L) (1) 220/380 (A) 6.88 kV

220 (L) 225/390 (A)220/380 (A) ± 10

Brazil 60 220 (L) (1) 220/380 (A) 13.8 kV (9)127 (L) (1) 127/220 (A) 11.2 kV

220/380 (A)127/220 (A)

Belgium 50 ± 3 220/380 (A) 220/380 (A) 15 kV + 5 (day)127/220 (A) 127/220 (A) 6 kV ± 10 (night)220 (F) 220 (F) 220/380 (A)

127/220 (A)220 (F)

Bolivia 50 ± 1 115/230 (H) 115/230 (H) 115/230 (H) (3) ± 5Cambodia 50 120/208 (A) 220/380 (A) 220/380 (A) (3) (9)

120 (L) 120/208 (A) 120/208 (A)Canada 60 ± 0.02 120/240 (K) 347/600 (A) 7.2/12.5 kV ± 4

480 (F) 347/600 (A) - 8.3240 (F) 120/208120/240 (K) 600 (F)120/208 (A) 480 (F)

240 (F)Chile 50 220 (L) (1) 220/380 (A) (1) 220/380 (A) (3) (9)China 50 220 (L) (1) 220/380 (A) 220/380 (A) (3) ± 7Colombia 60 ± 1 120/240 (G) 120/240 (G) 13.2 kV ± 10

120 (L) 120 (L) 120/240 (G)Costa Rica 60 120 (L) (1) 120/240 (K) 120/240 (G) (3) (9)

120 (L) (1)Czechoslovakia 50 ± 0.1 230/380 (A) 220/380 (A) 22 kV ± 10

220 (L) 220 (L) 15 kV6 kV3 kV220/380 (A)

Denmark 50 ± 0.4 220/380 (A) 220/380 (A) 30 kV ± 10220 (L) 220 (L) 10 kV

220/380 (A)Egypt (AR) 50 ± 1 220/380 (A) 220/380 (A) 11 kV + 10

220 (L) 220 (L) 6.6 kV220/380 (A)

Finland 50 ± 0.1 220 (L) (1) 220/380 (A) 380/660 (A) ± 10500 (B)220/380 (A) (D)

France 50 ± 1 230/400 (A) 230/400 (A) ± 10220/380 (A) 220/380 (A) 20 kV220 (L) 220/380 (D) 15 kV

230/400127/220 (E) 380 (B) 380 (B)127 (L) 220/380 (A) (D)

Germany 50 ± 0.3 220/380 (A) 220/380 (A) 20 kV ± 10Ex-DRG 220 (L) 220 (L) 10 kV

220/380 (A)

Ex-FRG 50 ± 0.3 220/380 (A) 220/380 (A) 10 kV ± 5220 (L) 220 (L) 6 kV127/220 (A) 380/660 (A)127 (L) 220/380 (A)

Greece 50 ± 1 220 (L) (1) 6.6 kV 22 kV ± 5220/380 (A) 20 kV

15 kV6.6 kV220/380 (A)

Hong Kong 50 ± 2 200/346 (A) 11 kV 11 kV ± 6(and Kowloon) 200 (L) (1) 200/346 (A) 200/346 (A)

220/380 (A) 220/380 (A) (3)200 (L)

* IEC 38 (1983).

table D1: survey of electricity supplies invarious countries around the world.Bracketed letters relate to the circuitdiagrams at the end of the table, whilebracketed numbers refer to the noteswhich follow the diagrams.

the adjoining table is extracted fromthe document "World ElectricitySupplies", fourth edition.

D2 - low-voltage service connections

D

1. low-voltage public distribution networks (continued)

1.1 low-voltage consumers (continued)


Hungary 50 ± 2 220/380 (A) 220/380 (A) 20 kV + 5 - 10220 (L) 220 (L)

10 kV220/380 (A)

Iceland 50 ± 0.1 220/380 (A) 220/380 (A) 220/380 (A) (3) (9)220 (L) 220 (L)

India (4)Bombay 50 ± 1 250/440 (A) 250/440 (A) 11 kW + 4

230 (L) 230 (L) 250/440 (A)

New Delhi 50 ± 3 230/400 (A) 230/400 (A) 11 kV + 6230 (L) 230 (L) 230/400 (A)

Roma- 50 ± 3 230/400 (A) 220/400 (A) 22 kV and 11 kV + 6krishnapuram (2) 25 230 (L) 230 (L) (9)

d.c. 230/460 (P) 230/460 (P) (9)Indonesia 50 ± 1 127/220 (A) 220/380 (A) 220/380 (A) (3) + 5

- 2 127/220 (A)Iran 50 ± 5 220 (L) (1) 220/380 (A) 20 kV + 15

11 kV231/400 (A)220/380 (A)

Iraq 50 220 (L) (1) 220/380 (A) 11 kV + 56.6 kV3 kV220/380 (A)

Ireland (Northern) 50 ± 0.4 230 (L) (1) 230/400 (A) 230/400 (A) (3) + 6220 (L) (1) 220/380 (A) 220/380 (A)

Ireland 50 220 (L) (1) 220/380 (A) 10 kV (9)(Republic of) 220/380 (A)Israel 50 ± 0.2 230/400 (A) 230/400 (A) 22 kV + 6

230 (L) 230 (L) 12.6 kV6.3 kV230/400 (A)

Italy 50 ± 0.4 220/380 (A) 220/380 (A) 20 kV ± 5 (urban)127/220 (E) 127/220 (E) 15 kV ± 10 (rural)220 (L) 10 kV

220/380 (A)220 (C)

Japan (East) (4) 50 ± 0.2 100/200 (K) 100/200 (H) (K) 6.6 kV ± 10(5) 100 (L) 100/200 (H)

200 (G) (J)

Japan (West) (4) 60 ± 0.1 105/210 (K) 105/210 (H) (K) 22 kV ± 10(5) 100/200 (K) 100/200 (K) 6.6 kV

100 (L) 100 (L) 105/210 (H)100/200 (H)

Korea (North) 60 + 0 220 (L) 220/380 (A) 220/380 (A) + 6.8- 5 - 13.6 (10)

Korea (South) 60 100 (L) 100/200 (K) (9) (9)Kuwait 50 240 (L) (1) 240/415 (A) 240/415 (A) (3) (9)Luxembourg 50 ± 0.5 220/380 (A) 220/380 (A) 20 kV ± 5 and ± 10

127/220 (A) 127/220 (A) 15 kV120/208 (A) 120/208 (A) 5 kV

Malaysia 50 ± 1.0 240 (L) (1) 240/415 (A) 240/415 (A) (3) + 5- 10

Mexico 60 ± 0.2 127/220 (A) 127/220 (A) 13.8 kV ± 6220 (L) 220 (L) 13.2 kV120 (M) 120 (M) 277/480 (A)

127/220 (B)Morocco 50 127/220 (A) 220/380 (A) 220/380 (A) (3) (9)

115/200 (A)Netherlands 50 ± 0.4 220/300 (A) 220/380 (A) 10 kV ± 6

220 (E) (L) 3 kV220/380 (A)

New Zealand 50 ± 1.5 230/400 (A) (E) 240/415 (A) (E) 11 kV ± 5230 (L) 230/400 (A) (E) 230/400 (A)240 (L) 230 (L) 240/415 (A)

240 (L) 440 (N) (6)Nigeria 50 ± 1 230 (L) (1) 230/400 (A) 15 kV ± 5

220 (L) (1) 220/380 (A) 11 kV230/400 (A)220/380 (A)

Norway 50 ± 0.2 230 (B) 220/380 (A) 20 kV ± 10230 (B) 10 kV

5 kV220/380 (A)230 (B)

Pakistan 50 230 (L) (1) 230/400 (A) 230/400 (A) (3) (9)230 (L)

Philippines 60 ± 0.16 110/220 (K) 13.8 kV 13.8 kV ± 54.16 kV 4.16 kV2.4 kV 2.4 kV110/220 (H) 440 V (B)

110/220 (H)

Manila 60 ± 5 240/120 (H) (K) 240/120 (H) (K) 20 kV ± 5240/120 (H) 240/120 (H) 6.24 kV

3.6 kV240/120 (H)

Peru 60 225 (B) (M) 225 (B) (M) 10 kV (9)6 kV225 (B)

table D1: survey of electricity supplies invarious countries around the world.Bracketed letters relate to the circuitdiagrams at the end of the table, whilebracketed numbers refer to the noteswhich follow the diagrams (continued).


D


Poland 50 ± 1 220 (L) (1) 220/380 (A) 15 kV ± 56 kV220/380 (A)

Portugal 50 ± 1 220/380 (A) 15 kV 15 kV ± 5220 (L) 5 kV 5 kV

220/380 (A) 220/380 (A)220 (L)

Rumania 50 ± 1 220 (L) (1) 220/380 (L) 20 kV ± 510 kV6 kV220/380 (A)

Saudi Arabia 60 ± 0.5 127/220 (A) 127/220 (A) 13.8 kV ± 5220/380 (A) 220/380 (A)

Singapore 50 ± 0.5 230/400 (A) 6.6 kV 22 kV ± 3230 (L) 230/400 (A) 6.6 kV

230/400 (A)Spain 50 ± 3 220/380 (A) (E) 220/380 (A) 15 kV ± 7

220 (L) 127/220 (A) 11 kV127/220 (A) (E) 220/380 (A)127 (L)

South Africa 50 ± 2.5 250/433 (A) (7) 11 kV 11 kV ± 525 (8) 230/400 (A) (7) 6.6 kV 6.6 kV

220/380 (A) 3.3 kV 3.3 kV220 (L) 250/433 (A) (7) 500 (B)230/400 (A) (7) 220/380 (A)220/380 (A)

Sweden 50 ± 0.2 220/380 (A) 220/380 (A) 20 kV ± 10220 (L) 220 (L) 10 kV

6 kV220/380 (A)

Syria 50 220 (L) (1) 220/380 (A) 220/380 (A) (3) (9)115 (L) (1) 220 (L) 115/200 (A)

115/200 (A)115 (L)

Taiwan 60 ± 4 220/380 (A) 220/380 (A) 22.8 kV ± 5 and ± 10220 (L) 110/220 (H) 11.4 kV110/220 (K) 220/380 (A)110 (L) 220 (H)

Tunisia 50 + 2 220/380 (A) 220/380 (A) 15 kV + 10220 (L) 220 (L) 10 kV

220/380 (A)Turkey 50 ± 2 220 (L) (1) 220/380 (A) 15 kV ± 10

6.3 kV220/380 (A)

United Kingdom 50 ± 1 240 (L) (1) 240/415 (A) 22 kV ± 611 kV6.6 kV3.3 kV240/415 (A)

U.S.A. (4)Charlotte 60 ± 0.06 120/240 (K) 265/460 (A) 14.4 kV + 5 - 2.5(North Carolina) 120/208 (A) 120/240 (K) 7.2 kV

120/208 (A) 2.4 kV575 (F)460 (F)240 (F)265/460 (A)120/240 (K)120/208 (A)

Detroit 60 ± 0.2 120/240 (K) 480 (F) 13.2 kV + 4 - 6.6(Michigan) 120/208 (A) 120/240 (H) 4.8 kV

120/208 (A) 4.16 kV480 (F)120/240 (H)120/208 (A)

Los Angeles 60 ± 0.2 120/240 (K) 4.8 kV 4.8 kV ± 5(California) 120/240 (G) 120/240 (G)

Miami 60 ± 0.3 120/240 (K) 120/240 (K) 13.2 kV ± 5(Florida) 120/208 (A) 120/240 (H) 2.4 kV

120/208 (A) 480/277 (A)120/240 (H)

New York 60 120/240 (K) 120/240 (K) 12.47 kV (9)(New York) 120/208 (A) 120/208 (A) 4.16 kV

240 (F) 277/480 (A)480 (F)

Pittsburgh 60 ± 0.03 120/240 (K) 265/460 (A) 13.2 kV ± 5 (lighting)(Pennsylvania) 120/240 (K) 11.5 kV ± 10 (power)

120/208 (A) 2.4 kV460 (F) 265/460 (A)230 (F) 120/208 (A)

460 (F)230 (F)

Portland 60 120/240 (K) 227/480 (A) 19.9 kV (9)(Oregon) 120/240 (K) 12 kV

120/208 (A) 7.2 kV480 (F) 2.4 kV240 (F) 277/480 (A)

120/208 (A)480 (F)240 (F)

table D1: survey of electricity supplies invarious countries around the world.Bracketed letters relate to the circuitdiagrams at the end of the table, whilebracketed numbers refer to the noteswhich follow the diagrams (continued).



D




San Francisco 60 ± 0.08 120/240 (K) 277/480 (A) 20.8 kV ± 5(California) 120/240 (K) 12 kV

4.16 kV277/480 (A)120/240 (G)

Toledo 60 ± 0.08 120/240 (K) 277/480 (C) 12.47 kV ± 5(Ohio) 120/208 (A) 120/240 (H) 7.2 kV

120/208 (K) 4.8 kV4.16 kV480 (F)277/480 (A)120/208 (A)

USSR (former) 50 220/380 (A) 220/380 (A) 220/380 (A) (3) (9)220 (L) 220 (L)127/220 (A)127 (L)

Viet-Nam 50 ± 0.1 220 (L) (1) 220/380 (A) 15 kV ± 10120 (L) (1) 120/208 (A) 220/380 (A)

Yugoslavia 50 220/380 (A) 220/380 (A) 10 kV (9)220 (L) 220 (L) 6.6 kV

220/380 (A)

table D1: survey of electricity supplies in various countries around the world.Bracketed letters relate to the circuit diagrams at the end of the table, while bracketednumbers refer to the notes which follow the diagrams.


Dcircuit diagrams*

(f) three-phase delta:three-wire

(e) two-phase star;three-wire:earthed neutral

(d) three-phase star;four-wire:non-earthed neutral

(a) three-phase star;four-wire:earthed neutral

(b) three-phase star:three-wire

(c) three-phase star;three-wire:earthed neutral

(g) three-phase delta;four-wire:earthed mid point of one phase

(j) three-phase open delta:earthed junction of phases

(h) three-phase open delta;four-wire:earthed mid point of one phase

(k) single-phase;three-wire:earthed mid point

(l) single-phase;two-wire:earthed end of phase

(m) single-phase;two-wire:unearthed

(n) single-wire:earthed return (swer)

(p) d.c.:three-wire: unearthed

V kV

* Windings (A) (B) (C) (D) and (F) may be transformer-secondary windings or alternator stator windings.

Notes(1) The supply to each house is normallysingle-phase using one line and one neutralconductor of systems (A) or (G).(2) Frequencies below 50 Hz and d.c.supplies are used in limited areas only. Theexamples given show the diversity ofpossibilities existing.(3) Information on higher voltage suppliesto factories is not available.(4) More than one area of country is givento illustrate the differences which exist.

(5) Frequency is 50 Hz (eastern area) and60 Hz (western area). The dividing line fromnorth to south passes through Shizuoka onHonshu Island.(6) Some remote areas are supplied via aSingle Wire Earthed Return (SWER) system.(7) A few towns only have this supply.(8) Refers to isolated mining districts.(9) Information not available.(10) Observed values.


D



residential and commercialconsumersThe function of a LV "mains" distributor(underground cable or overhead line) is toprovide service connections to a number ofconsumers along its route.The current-rating requirements ofdistributors are estimated from the number ofconsumers to be connected and an averagedemand per consumer.The two principal limiting parameters of adistributor are:c the maximum current which it is capable ofcarrying indefinitely, andc the maximum length of cable which, whencarrying its maximum current, will not exceedthe statutory voltage-drop limit.

These constraints mean that the magnitudeof loads which power-supply organizationsare willing to connect to their LV distributionmains, is necessarily restricted.For the range of LV systems mentioned in thesecond paragraph of this sub-clause (1.1) viz:120 V single phase to 240/415 V 3-phase,typical maximum permitted loads connectedto a LV distributor might* be:

system assumed max. permitted current kVAper consumer service

120 V 1-phase 2-wire 60 A 7.2120/240 V 1-phase 3-wire 60 A 14.4120/208 V 3-phase 4-wire 60 A 22220/380 V 3-phase 4-wire 120 A 80230/400 V 3-phase 4-wire 120 A 83240/415 V 3-phase 4-wire 120 A 86table D2.

* The table D2 values shown are indicative only, being (arbitrarily) based on 60 A maximum service currents for the first threesystems, since smaller voltage drops are allowed at these lower voltages, for a given percentage statutory limit.The second group of systems is (again, arbitrarily) based on a maximum permitted service current of 120 A.

Practices vary considerably from one power-supply organization to another, and no"standardized" values can be given.Factors to be considered include:c the size of an existing distributor to whichthe new load is to be connected,c the total load already connected to thedistributor,c the location along the distributor of theproposed new load, i.e. close to thesubstation, or near the remote end of thedistributor, etc.

In short, each case must be examinedindividually.The load levels listed above are adequate forall normal domestic consumers, and will besufficient for the installations of manyadministrative, commercial and similarbuildings.

medium-size and small industrialconsumers (with dedicated LVlines direct from a public-supplyHV/LV substation)Medium and small industrial consumers canalso be satisfactorily supplied at low-voltage.For loads which exceed the maximumpermitted limit for a service from a distributor,a dedicated cable can usually be providedfrom the LV distribution fuse- (or switch-)board from which the mains distributorsemanate, in the power-supply authoritysubstation.In principle, the upper load limit which can besupplied by this means is restricted only bythe available spare transformer capacity inthe substation.In practice, however:c large loads (e.g. > 300 kVA) requirecorrespondingly large cables, so that, unlessthe load centre is close to the substation,this method can be economicallyunfavourable,c many power-supply organizations prefer tosupply loads exceeding 200 kVA (this figurevaries with different suppliers) at high voltage.

For these reasons, dedicated supply lines atLV are generally applied (at 220/380 V to240/415 V) to a load range of 80 kVA to250 kVA.Consumers normally supplied at low voltageinclude:c domestic dwellings,c shops and commercial buildings,c small factories, workshops and fillingstations,c restaurants,c farms, etc.


D

1.2. LV distribution networks

in cities and large towns, standard-sized LV distribution cables form anetwork through link boxes. Somelinks are removed, so that each(fused) distributor leaving asubstation forms a branched open-ended radial system, as shown infigure D3.

In European countries the standard3-phase 4-wire distribution voltage levels are220/380 V, 230/400 V, or 240/415 V. Manycountries are currently converting their LVsystems to the latest IEC standard of230/400 V nominal (IEC 38-1983). The targetdate for completion is the year 2003. Mediumto large-sized towns and cities haveunderground cable distribution systems.HV/LV distribution substations, mutuallyspaced at approximately 500-600 metres, aretypically equipped with:c a 3-or 4-way HV switchboard, often madeup of incoming and outgoing load-breakswitches forming part of a ring main, and oneor two HV circuit breakers or combined fuse/load-break switches for the transformercircuits,c one or two 1.000 kVA HV/LV transformers,c one or two (coupled) 6-or 8-way LV3-phase 4-wire distribution fuse boards, ormoulded-case circuit breaker boards, controland protect outgoing 4-core distributioncables, generally referred to as “distributors".The output from a transformer is connectedto the LV busbars via a load-break switch, orsimply through isolating links.

In densely-loaded areas, a standard size ofdistributor is laid to form a network, with(generally) one cable along each pavementand 4-way link boxes located in manholes atstreet corners, where two cables cross.Recent trends are towards weather-proofcabinets above ground level, either against awall, or where possible, flush-mounted in thewall.Links are inserted in such a way thatdistributors form radial circuits from thesubstation with open-ended branches (seefigure D3). Where a link box unites adistributor from one substation with that froma neighbouring substation, the phase linksare omitted or replaced by fuses, but theneutral link remains in place.This arrangement provides a very flexiblesystem in which a complete substation canbe taken out of service, while the areanormally supplied from it is fed from linkboxes of the surrounding substations.Moreover, short lengths of distributor(between two link boxes) can be isolated forfault-location and repair.

4-waylink box

HV/LVsubstation

service cable

phase linksremoved

fig. D3: showing one of several ways in which a LV distribution network may be arrangedfor radial branched-distributor operation, by removing (phase) links.


D


1.2 LV distribution networks (continued)

in less-densely loaded urban areas amore-economic system of taperedradial distribution is commonly used,in which conductors of reduced sizeare installed as the distance from asubstation increases.

Where the load density requires it, thesubstations are more closely spaced, andtransformers up to 1.500 kVA are sometimesnecessary.Other forms of urban LV network, based onfree-standing LV distribution pillars, placedabove ground at strategic points in thenetwork, are widely used in areas of lowerload density. This scheme exploits theprinciple of tapered radial distributors in whichthe distribution cable conductor size isreduced as the number of consumersdownstream diminish with distance from thesubstation.In this scheme a number of large-sectionedLV radial feeders from the distribution boardin the substation each supply the busbars ofa distribution pillar, from which smallerdistributors supply consumers immediatelysurrounding the pillar.Distribution in market towns, villages andrural areas generally has, for many years,been based on bare copper conductorssupported on wooden, concrete or steelpoles, and supplied from pole-mounted orground-mounted transformers.

In recent years, LV insulated conductors,twisted to form a two-core or 4-core self-supporting cable for overhead use, havebeen developed, and are considered to besafer and visually more acceptable than barecopper lines.This is particularly so when the conductorsare fixed to walls (e.g. under-eaves wiring)where they are hardly noticeable.As a matter of interest, similar principles havebeen applied at higher voltages, and self-supporting “bundled” insulated conductors forHV overhead installations are now availablefor operation at 24 kV.Where more than one substation supplies avillage, arrangements are made at poles onwhich the LV lines from different substationsmeet, to interconnect corresponding phasesat times of emergency. The neutralconductors are permanently connected.

In Europe, each public-supply distributionsubstation is able to supply at LV an areacorresponding to a radius of approximately300 metres from the substation.North and Central American practice differsfundamentally from that in Europe, in that LVnetworks are practically nonexistent, and3-phase supplies to domestic premises inresidential areas are rare.The distribution is effectively carried out athigh voltage in a way, which again differsfrom standard European practices. The HVsystem is, in fact, a 3-phase 4-wire systemfrom which single-phase distributors (phaseand neutral conductors) supply numeroussingle-phase transformers, the secondarywindings of which are centre-tapped toproduce 120/240 V single-phase 3-wiresupplies. The central conductors provide theLV neutrals, which, together with the HVneutral conductors, are solidly earthed atintervals along their lengths.Each HV/LV transformer normally suppliesone or several premises directly from thetransformer position by radial service cable(s)or by overhead line(s).

improved methods using insulatedtwisted conductors to form a pole-mounted aerial cable are nowstandard practice in many countries.

in Europe, each public-supplydistribution substation is able tosupply at LV an area correspondingto a radius of approximately300 metres from the substation.North and Central American systemsof distribution consist of a HVnetwork from which numerous(small) HV/LV transformers eachsupply one or several consumers, bydirect service cable (or line) from thetransformer location.


DMany other systems exist in these countries,but the one described appears to be the mostcommon.Figure D4 shows the main features of the twosystems.

2.4 kV / 120-240 V1 ph - 3 wiredistributiontransformer

N

2

N

N

1

N

3

N

N

2

1 2 3

N

13.8 kV / 2.4-4.16 kV

each HV/LV transformer shownrepresents many similar units

main 3 ph and neutralHV distributor

N3

2

1

21

3 phHV / 230/400 V4-wire distributiontransformer

resistor replacedby a Petersen coilon O/H line systemsin some countries

1 2 3N

1 phHV / 230 Vservice transformerto isolated consumer(s)(rural supplies)

Ph N

for primary voltages > 72.5 kV(see note)primary winding may be :– delta– earthed star– earthed zigzagdepending on the country concerned

tertiary deltanormally (not always)used if the primarywinding is not delta

with on-loadtap changer

LV distribution network

HV (1)

HV (2)

(1): 132 kV for example(2): 11 kV for example

fig. D4: widely-used American and European-type systems.

Note: at primary voltages greater than72.5 kV in bulk-supply substations, it iscommon practice in some Europeancountries to use an earthed-star primarywinding and a delta secondary winding. Theneutral point on the secondary side is thenprovided by a zigzag earthing reactor, the starpoint of which is connected to earth through aresistor.Frequently, the earthing reactor has asecondary winding to provide LV 3-phasesupplies for the substation. It is then referredto as an “earthing transformer”.


D


1.3 the consumer-service connection

service components and meteringequipment were formerly installedinside a consumer's building. Themodern tendency is to locate theseitems outside in a weatherproofcabinet.

In the past, an underground cable service orthe wall-mounted insulated conductors froman overhead line service, invariablyterminated inside the consumer's premises,where the cable-end sealing box, the supply-authority fuses (inaccessible to theconsumer) and meters were installed.A more recent trend is (as far as possible) tolocate these service components in aweatherproof housing outside the building.The supply-authority/consumer interface isoften at the outgoing terminals of the meter(s)or, in some cases, at the outgoing terminalsof the installation main circuit breaker(depending on local practices) to whichconnection is made by supply-authoritypersonnel, following a satisfactory test andinspection of the installation.A typical arrangement is shown in figure D5.

fig. D5: typical service arrangement for TT-earthed systems.

A = Service cable tee-jointF = Supply authority fusesC = Metering equipmentS = Isolating linkDB = Installation main circuit breaker

A MCCB which incorporates a sensitiveresidual-current earth-fault protective featureis mandatory at the origin of any LVinstallation forming part of a TT earthingsystem. The reason for this feature andrelated leakage-current tripping levels arediscussed in Clause 3 of Chapter G.A further reason for this MCCB is that theconsumer cannot exceed his (contractual)declared maximum load, since the overloadtrip setting, which is sealed by the supplyauthority, will cut off supply above thedeclared value. Closing and tripping of theMCCB is freely available to the consumer, sothat if the MCCB is inadvertently tripped onoverload, or due to an appliance fault,supplies can be quickly restored followingcorrection of the anomaly.

In view of the inconvenience to both themeter reader and consumer, the location ofmeters is nowadays generally outside thepremises, either:c in a free-standing pillar-type housing asshown in figures D6 and D7,c in a space inside a building, but with cabletermination and supply authority’s fuseslocated in a flush-mounted weatherproofcabinet accessible from the public way, asshown in figure D8.

LV consumers are normally suppliedaccording to the TN or TT system, asdescribed in chapters F and G. Theinstallation main circuit breaker for aTT supply must include a residual-current earth-leakage protectivedevice. For a TN service, overcurrentprotection by circuit breaker orswitch-fuse is required.


D

fig. D6: typical rural-type installation.

In this kind of installation it is often necessaryto place the main installation circuit breakersome distance from the point of utilization,e.g. saw-mills, pumping stations, etc.

fig. D7: semi-urban installations (shopping precincts, etc.).

The main installation CB is located in theconsumer's premises in cases where it is setto trip if the declared kVA load demand isexceeded.

fig. D8: town centre installations.

The service cable terminates in a flush-mounted wall cabinet which contains theisolating fuse links, accessible from the publicway. This method is preferred for estheticreasons, when the consumer can provide asuitable metering and main-switch location.


D


1.3 the consumer-service connection (continued)

c for private domestic consumers, theequipment shown in the cabinet in figure D5is installed in a weatherproof cabinetmounted vertically on a metal frame in thefront garden, or flush-mounted in theboundary wall, and accessible to authorizedpersonnel from the pavement. Figure D9shows the general arrangement, in whichremovable fuse links provide the means ofisolation.Experiments are now well-advanced in thefield of electronic metering; reading, andrecording on magnetic cards is now possible,using information technology (IT) techniques,and it is confidently predicted that, in additionto remote reading and recording, themodification of tariff structures for a givenmeter will be possible from a central controllocation, in areas where it is economicallyjustified.

isolation by fuse links

overhead line LV distributor

supply authority/consumer interface

meter cabinet

metermaininstallationcircuitbreaker

installation

service cable

fig. D9: typical LV service arrangement for domestic consumers.


D

1.4 quality of supply voltage

The quality of the LV network supply voltagein its widest sense implies:c compliance with statutory limits ofmagnitude and frequency,c freedom from continual fluctuation withinthose limits,c uninterrupted power supply, except forscheduled maintenance shutdowns, or as aresult of system faults or other emergencies,c preservation of a near-sinusoidal waveform.In this Sub-clause the maintenance ofvoltage magnitude only will be discussed,the remaining subjects are covered inClause 2 of chapter F.In most countries, power-supply authoritieshave a statutory obligation to maintain thelevel of voltage at the service position ofconsumers within the limits of ± 5% (or insome cases ± 6% or more-see table D1) ofthe declared nominal value.Again, IEC and most national standardsrecommend that LV appliances be designedand tested to perform satisfactorily within thelimits of ± 10% of nominal voltage. Thisleaves a margin, under the worst conditions(of minus 5% at the service position, forexample) of 5% allowable voltage drop inthe installation wiring.The voltage drops in a typical distributionsystem occur as follows:the voltage at the HV terminals of a HV/LVtransformer is normally maintained within a± 2% band by the action of automatic on-load tapchangers of the transformers atbulk-supply substations, which feed the HVnetwork from a higher-voltage sub-transmission system.

If the HV/LV transformer is in a locationclose to a bulk-supply substation, the ± 2%voltage band may be centred on a voltagelevel which is higher than the nominal HVvalue. For example, the voltage could be20.5 kV± 2% on a 20 kV system. In this case, theHV/LV distribution transformer should haveits HV off-circuit tapping switch selected tothe+ 2.5% tap position.Conversely, at locations remote from bulk-supply substations a value of 19.5 kV ± 2%is possible, in which case the off-circuittapping switch should be selected to the- 5% position.The different levels of voltage in a system arenormal, and depend on the system power-flow pattern. Moreover, these voltagedifferences are the reason for the term“nominal” when referring to the systemvoltage.

practical applicationWith the HV/LV transformer correctlyselected at its off-circuit tapping switch, anunloaded transformer output voltage will beheld within a band of± 2% of its no-load voltage output.To ensure that the transformer can maintainthe necessary voltage level when fullyloaded, the output voltage at no-load mustbe as high as possible without exceedingthe upper + 5% limit (adopted for thisexample).In present-day practice, the winding ratiosgenerally give an output voltage of about

an adequate level of voltage at theconsumers supply-service terminalsis essential for satisfactory operationof equipment and appliances.Practical values of current, andresulting voltage drops in a typical LVsystem, show the importance ofmaintaining a high Power Factor as ameans of reducing voltage drop.

104% at no-load*, when nominal voltage isapplied at HV, or is corrected by the tappingswitch, as described above. This wouldresult in a voltage band of 102% to 106% inthe present case.A typical LV distribution transformer has ashort-circuit reactance voltage of 5%. If it isassumed that its resistance voltage is onetenth of this value, then the voltage dropwithin the transformer when supplying fullload at 0.8 power factor lagging, will be:V% drop = R% cos ø + X% sin ø

= 0.5 x 0.8 + 5 x 0.6= 0.4 + 3 = 3.4%

The voltage band at the output terminals ofthe fully-loaded transformer will therefore be(102 - 3.4) = 98.6% to (106 - 3.4) = 102.6%.The maximum allowable voltage drop alonga distributor is therefore 98.6 - 95 = 3.6%.This means, in practical terms, that amedium-sized 230/400 V 3-phase 4-wiredistribution cable of 240 mm2 copperconductors would be able to supply a totalload of 292 kVA at 0.8 PF lagging,distributed evenly over 306 metres of thedistributor. Alternatively, the same load at thepremises of a single consumer could besupplied at a distance of 153 metres fromthe transformer, for the same volt-drop, andso on...As a matter of interest, the maximum ratingof the cable, based on calculations derivedfrom IEC 287 (1982) is 290 kVA, and so the3.6% voltage margin is not unduly restrictive,i.e. the cable can be fully loaded fordistances normally required in LV distributionsystems.Furthermore, 0.8 PF lagging is appropriateto industrial loads. In mixed semi-industrialareas 0.85 is a more common value, while0.9 is generally used for calculationsconcerning residential areas, so that thevolt-drop noted above may be considered asa “worst case” example.* Transformers designed for the 230/400 V IEC standardwill have a no-load output of 420 V, i.e. 105% of thenominal voltage.


D

2. tariffs and metering

tariffs and metering

No attempt will be made in this guide todiscuss particular tariffs, since there appearsto be as many different tariff structuresaround the world as there are distributionauthorities.Some tariffs are very complicated in detail butcertain elements are basic to all of them andare aimed at encouraging consumers tomanage their power consumption in a waywhich reduces the cost to the supply authorityof generation, transmission and distribution.The two predominant ways in which the costof supplying power to consumers can bereduced, are:c reduction of power losses in the generation,transmission and distribution of electricalenergy. In principle the lowest losses in apower system are attained when all parts ofthe system operate at unity power factor,c reduction of the peak power demand, whileincreasing the demand at low-load periods,thereby exploiting the generating plant morefully, and minimizing plant redundancy.

reduction of lossesAlthough the ideal condition noted in the firstpossibility mentioned above cannot berealized in practice, many tariff structures arebased partly on kVA demand, as well as onkWh consumed.Since, for a given kW loading, the minimumvalue of kVA occurs at unity power factor, theconsumer can minimize billing costs by takingsteps to improve the power factor of the load(as discussed in Chapter E).The kVA demand generally used for tariffpurposes is the maximum average kVAdemand occurring during each billing period,and is based on average kVA demands, overfixed periods (generally 10, 30 or 60 minuteperiods) and selecting the highest of thesevalues.The principle is described below in "principleof kVA maximum-demand metering".

reduction of peak power demandThe second aim, i.e. that of reducing peakpower demands, while increasing demand atlow-load periods, has resulted in tariffs whichoffer substantial reduction in the cost ofenergy at:c certain hours during the 24-hour day,c certain periods of the year.The simplest example is that of a domesticconsumer with a storage-type water heater(or storage-type space heater, etc.). Themeter has two digital registers, one of whichoperates during the day and the other(switched over by a timing device) operatesduring the night.A contactor, operated by the same timingdevice, closes the circuit of the water heater,the consumption of which is then indicated onthe register to which the cheaper rate applies.The heater can be switched on and off at anytime during the day if required, but will thenbe metered at the normal rate.Large industrial consumers may have 3 or 4rates which apply at different periods during a24-hour interval, and a similar number fordifferent periods of the year.In such schemes the ratio of cost per kWhduring a period of peak demand for the year,and that for the lowest-load period of theyear, may be as much as 10: 1.

metersIt will be appreciated that high-qualityinstruments and devices are necessary toimplement this kind of metering, when usingclassical electro-mechanical equipment.Recent developments in electronic meteringand micro-processors, together with remoteripple-control* from a supply-authority controlcentre (to change peak-period timingthroughout the year, etc.) are nowoperational, and facilitate considerably theapplication of the principles discussed.* Ripple control is a system of signalling in which a voice-frequency current (commonly at 175 Hz) is injected into theLV mains at appropriate substations. The signal is injectedas coded impulses, and relays which are tuned to the signalfrequency and which recognize the particular code willoperate to initiate a required function. In this way, up to 960discrete control signals are available.


DIn most countries, certain tariffs, as notedabove, are partly based on kVA demand, inaddition to the kWh consumption, during thebilling periods (often 3-monthly intervals).The maximum demand registered by themeter to be described, is, in fact, a maximum(i.e. the highest) average kVA demand

registered for succeeding periods during thebilling interval.Figure D10 shows a typical kVA demandcurve over a period of two hours divided intosucceeding periods of 10 minutes. The metermeasures the average value of kVA duringeach of these 10 minute periods.

0 1 2 hrstime

kVA

maximum average value during the 2 hour interval

average valuesfor 10 minuteperiods

fig. D10: maximum average value of kVAover an interval of 2 hours.

principle of kVA maximum-demand metering

A kVAh meter is similar in all essentials to akWh meter but the current and voltage phaserelationship has been modified so that iteffectively measures kVAh (kilo-volt-amp-hours).Furthermore, instead of having a set ofdecade counter dials, as in the case of aconventional kWh meter, this instrument hasa rotating pointer.When the pointer turns it is measuring kVAhand pushing a red indicator before it.At the end of 10 minutes the pointer will havemoved part way round the dial (it is designedso that it can never complete one revolutionin 10 minutes) and is then electrically reset tothe zero position, to start another 10 minuteperiod.The red indicator remains at the positionreached by the measuring pointer, and thatposition, corresponds to the number of kVAh(kilo-volt-ampere-hours) taken by the load in10 minutes.Instead of the dial being marked in kilo-VA-hours at that point however it can be markedin units of average kVA.The following figures will clarify the matter.Supposing the point at which the red indicatorreached corresponds to 5 kVAh. It is knownthat a varying amount of kVA of apparentpower has been flowing for 10 minutes, i.e.1/6 hour.If now, the 5 kVAh is divided by the number ofhours, then the average kVA for the period isobtained.In this case the average kVA for the periodwill be:

5 x 1 = 5 x 6 = 30 kVA 1/6Every point around the dial will be similarlymarked i.e. the figure for average kVA will be6 times greater than the kVAh value at anygiven point.Similar reasoning can be applied to any otherreset-time interval.

At the end of the billing period, the redindicator will be at the maximum of all theaverage values occurring in the billing period.The red indicator will be reset to zero at thebeginning of each billing period.Electro-mechanical meters of the kinddescribed are rapidly being replaced byelectronic instruments. The basic measuringprinciples on which these electronic metersdepend however, are the same as thosedescribed above.

power factor improvement

E

Power factor improvementand harmonic filtering

1. Power factor improvement1.1 the nature of reactive energy E11.2 plant and appliances requiring reactive current E21.3 the power factor E21.4 tan ϕ E31.5 practical measurement of power factor E41.6 practical values of power factor E4

2. Why improve the power factor?2.1 reduction in the cost of electricity E52.2 technical/economic optimization E5

3. How to improve the power factor3.1 theoretical principles E63.2 by using what equipment? E73.3 the choice between a fixed or automatically-regulated bank of capacitors E8

4. Where to install correction capacitors4.1 global compensation E94.2 compensation by sector E94.3 individual compensation E10

5. How to decide the optimum level of compensation5.1 general method E115.2 simplified method E115.3 method based on the avoidance of tariff penalties E135.4 method based on reduction of declared maximum apparent power (kVA) E13

6. Compensation at the terminals of a transformer6.1 compensation to increase the available active power output E146.2 compensation of reactive energy absorbed by the transformer E15

7. Compensation at the terminals of an inductionmotor7.1 connection of a capacitor bank and protection settings E177.2 how self-excitation of an induction motor can be avoided E18

8. Example of an installation before E20and after power-factor correction9. The effects of harmonics on the ratingof a capacitor bank9.1 problems arising from power-system harmonics E219.2 possible solutions E219.3 choosing the optimum solution E229.4 possible effects of power-factor-correction capacitors E23on the power-supply system

10. Implementation of the capacitor banks10.1 capacitor element E2410.2 choice of protection, control devices, and connecting cables E25

power factor improvement - E1

E

1. power factor improvement

alternating current systems supplytwo forms of energy:c "active" energy measured inkilowatt hours (kWh) which isconverted into mechanical work,heat, light, etc.c "reactive" energy, which againtakes two forms:v "reactive" energy required byinductive circuits (transformers,motors, etc.),v "reactive" energy required bycapacitive circuits (cablecapacitance, power capacitors, etc).

All inductive (i.e. electromagnetic) machinesand devices that operate on a.c. systemsconvert electrical energy from the power-system generators into mechanical work andheat. This energy is measured by kWhmeters, and is referred to as "active" or"wattful" energy. In order to perform thisconversion, magnetic fields have to beestablished in the machines, and these fieldsare associated with another form of energy tobe supplied from the power system, knownas "reactive" or "wattless" energy.

The reason for this is that inductive plantcyclically absorbs energy from the system(during the build-up of the magnetic fields)and re-injects that energy into the system(during the collapse of the magnetic fields)twice in every power-frequency cycle. Theeffect on generator rotors is to (tend to) slowthem during one part of the cycle and toaccelerate them during another part of thecycle. The pulsating torque is stricly true onlyfor single-phase alternators. In three-phasealternators the effect is mutually cancelled inthe three phases, since, at any instant, thereactive energy supplied on one (or two)phase(s) is equal to the reactive energy beingreturned on the other two (or one) phase(s) ofa balanced system. The nett result is zeroaverage load on the generators, i.e. thereactive current is "wattless".An exactly similar phenomenon occurs withshunt capacitive elements in a power system,such as cable capacitance or banks of powercapacitors, etc. In this case, energy is storedelectrostatically. The cyclic charging anddischarging of capacitive plant reacts on thegenerators of the system in the same manneras that described above for inductive plant,but the current flow to and from capacitiveplant is in exact phase opposition to that ofthe inductive plant.This feature is the basis on which power-factor improvement schemes depend.

It should be noted that while this "wattless"current (more accurately, the wattlesscomponent of a load current) does not drawpower from the system, it does cause powerlosses in transmission and distributionsystems by heating the conductors.

In practical power systems, wattlesscomponents of load currents are invariablyinductive, while the impedances oftransmission and distribution systems arepredominantly inductively reactive. Thecombination of inductive current passingthrough an inductive reactance produces theworst possible conditions of voltage drop(i.e. in direct phase opposition to the systemvoltage).

1.1 the nature of reactive energy

For these reasons, viz:c transmission power losses andc voltage drop.The power-supply authorities reduce theamount of wattless (inductive) current asmuch as possible.Wattless (capacitive) currents have thereverse effect on voltage levels and producevoltage-rises in power systems.The power (kW) associated with "active"energy is usually represented by the letter P.The reactive power (kvar) is representedby Q. Inductively-reactive power isconventionally positive (+ Q) whilecapacitively-reactive power is shown as anegative quantity (- Q).

Sub-clause 1.3 shows the relationshipbetween P, Q, and S.S represents kVA of "apparent" power.Figure E1 shows that the kVA of apparentpower is the vector sum of the kW of activepower plus the kvar of reactive power.

Q(kvar)

S(kVA)

P(kW)

fig. E1: an electric motor requires activepower P and reactive power Q from thepower system.

E2 - power factor improvement

E

1. power factor improvement (continued)

1.2 plant and appliances requiring reactive current

items of plant which require reactiveenergy.

All a.c. plant and appliances that includeelectromagnetic devices, or depend onmagnetically-coupled windings, require somedegree of reactive current to create magneticflux.The most common items in this class aretransformers and reactors, motors anddischarge lamps (i.e. the ballasts of).The proportion of reactive power (kvar) withrespect to active power (kW) when an item ofplant is fully loaded varies according to theitem concerned being:c 65-75% for asynchronous motors,c 5-10% for transformers.The power factor at which a synchronousmotor operates may be changed, byadjustment of the excitation current. Thesemachines can be made to operate at lagging(underexcited) or leading (overexcited) powerfactors. In the latter condition, a synchronousmotor is sometimes referred to as a"synchronous condenser".Before capacitor technology had developedsufficiently to guarantee the high standard ofreliability of modern capacitors, synchronouscondensers were widely used ontransmission systems to provide reactive-power compensation, for optimumtransmission-line performance underchanging load conditions.

fig. E2: power consuming items that alsorequire reactive energy.

1.3 the power factor

the power factor is the ratio of kW tokVA. The closer the power factorapproaches its maximum possiblevalue of 1, the greater the benefit toconsumer and supplier.

PF = P (kW) ≈ cos ϕ S (kVA)P = active powerS = apparent power

definition of power factorThe power factor of a load, which may be asingle power-consuming item, or a number ofitems (for example an entire installation), isgiven by the ratio of P/S i.e. kW divided bykVA at any given moment.The value of a power factor will range from0 to 1.The power diagram of figure E3 shows thatthe ratio mentioned above gives the cosinevalue for the angular displacement betweenthe kW vector and the kVA vector.Conventionally, this angle is given thesymbol ϕ, in which case power factor = cos ϕ.

The accuracy of this equivalence depends onan absence of harmonic currents andvoltages on the system. It is generallyassumed that these effects are small, so thatcos ϕ and power factor are considered to beexact equivalents, for all practical purposes.A power factor close to unity means that thereactive energy is small compared with theactive energy, while a low value of powerfactor indicates the opposite condition.


Epower vector diagramActive power P (in kW)c single phase (1 phase and neutral):P = VI cos ϕc single phase (phase to phase):P = UI cos ϕc three phase (3 wires or 3 wires + neutral):P = e UI cos ϕ*

Reactive power Q (in kvar)c single phase (1 phase and neutral):Q = VI sin ϕc single phase (phase to phase): Q = UI sin ϕc three phase (3 wires or 3 wires + neutral):Q = e UI sin ϕ*Apparent power S (in kVA)c single phase (1 phase and neutral): S = VIc single phase (phase to phase): S = UIc three phase (3 wires or 3 wires + neutral):S = e UI*where:V: voltage between phase and neutralU: voltage between phases

* for balanced and near-balanced loads on4-wire systems.

P2+ Q

2S =

fig. E3: power diagram.P: active powerQ: reactive powerS: apparent power

fig. E4: current and voltage vectordiagram per phase.

current and voltage vectors, andderivation of the power diagramThe power "vector" diagram is a usefulartifice, derived directly from the true rotatingvector diagram of currents and voltage, asfollows:The power-system voltages are taken as thereference quantities, and one phase only isconsidered on the assumption of balanced3-phase loading.The reference phase voltage (V) isco-incident with the horizontal axis, and thecurrent (I) of that phase will, for practically allpower-system loads, lag the voltage by anangle ϕ.The component of I which is in phase with Vis the wattful component of I and is equal toI cos ϕ, while VI cos ϕ equals the activepower (in kV) in the circuit, if V is expressedin kV.The component of I which lags 90 degreesbehind V is the wattless component of I andis equal to I sin ϕ, while VI sin ϕ equals thereactive power (in kvar) in the circuit, if V isexpressed in kV.If the vector I is multiplied by V, expressed inkV, then VI equals the apparent power(in kVA) for the circuit.The above kW, kvar and kVA values perphase, when multiplied by 3, can thereforeconveniently represent the relationships ofkVA, kW, kvar and power factor for a total3-phase load, as shown in figure E3.

1.4 tan ϕ

tan ϕ = Q (kvar) P (kW)

Some electricity tariffs are partly based onthis factor, which shows the amount ofreactive power supplied per kW. A low valueof tan ϕ corresponds to a high power factorand to a favourable consumer bill.

power quantities kA, kVA and kvarare double-frequency functions andcannot be represented on a simplevector diagram. A static diagram forthese quantities (figure E3) can beobtained, however, to provide avisual representation, with the aid ofthe true vector diagram of currentcomponents and one phase-voltage(figure E4). Since, in the diagram, thepower quantities have direction andmagnitude, they are referred to as"vectors" for convenience.

kvarkVA

kWϕ P

SQ

Q = VI sin ϕ (kvar)

S = VI (kVA)

Vϕ

P = VI cos ϕ (kW)


E

1. power factor improvement (continued)

1.5 practical measurement of power factor

The power factor (or cos ϕ) can bemeasured, either:c by a direct-reading cos ϕ meter for aninstantaneous value, orc a recording var meter, which allows arecord over a period of time to be obtained, ofcurrent, voltage and power factor. Readingstaken over an extended period provide auseful means of estimating an average valueof power factor for an installation.

1.6 practical values of power factor

an example of power calculations

type of apparent power active power reactive powercircuit S (kVA) P (kW) Q (kvar)single-phase (phase and neutral) S = VI P = VI cos ϕ Q = VI sin ϕsingle-phase (phase to phase) S = UI P = UI cos ϕ Q = UI sin ϕexample 5 kW of load 10 kVA 5 kW 8.7 kvar

cos ϕ = 0.5three phase 3-wires or 3-wires + neutral S = e UI P = e UI cos ϕ Q = e UI sin ϕexample motor Pn = 51 kW 65 kVA 56 kW 33 kvar

cos ϕ = 0.86ρ = 0.91 (motor efficiency)

table E5: example in the calculation of active and reactive power.

The calculations for the three-phase exampleabove are as follows:Pn = delivered shaft power = 51 kWP = active power consumed =Pn = 51 = 56 kW ρ 0.91S = apparent power = P = P = 65 kVAcos ϕ 0.86so that, on referring to Table E20 or using apocket calculator, the value of tan ϕcorresponding to a cos ϕ of 0.86 is found tobe 0.59Q = P tan ϕ = 56 x 0.59 = 33 kvar.Alternatively

S2- P

2Q = = 65

2- 56

2= 33 kvar

S = 65 kVA

P = 56 kW

Q = 33 kvar

ϕ

fig. E6: calculation power diagram.

in addition to other informationTable E20 gives correspondingcosine and tangent values for givenangles.

average power factor values for the most commonly-used plant,equipment and appliances

plant and appliances cos ϕ tan ϕc common loaded at 0% 0.17 5.80induction 25% 0.55 1.52motor 50% 0.73 0.94

75% 0.80 0.75100% 0.85 0.62

c incandescent lamps 1.0 0c fluorescent lamps (uncompensated) 0.5 1.73c fluorescent lamps (compensated) 0.93 0.39c discharge lamps 0.4 to 0.6 2.29 to 1.33c ovens using resistance elements 1.0 0c induction heating ovens (compensated) 0.85 0.62c dielectric type heating ovens 0.85 0.62c resistance-type soldering machines 0.8 to 0.9 0.75 to 0.48c fixed 1-phase arc-welding set 0.5 1.73c arc-welding motor-generating set 0.7 to 0.9 1.02 to 0.48c arc-welding transformer-rectifier set 0.7 to 0.8 1.02 to 0.75c arc furnace 0.8 0.75

table E7: values of cos ϕ and tan ϕ for commonly-used plant and equipment.


E

2. why improve the power factor?

2.1 reduction in the cost of electricity

an improvement of the power factorof an installation presents severaltechnical and economic advantages,notably in the reduction of electricitybills.

Good management in the consumption ofreactive energy brings with it the followingeconomic advantages.These notes are based on an actual tariffstructure of a kind commonly applied inEurope, designed to encourage consumers tominimize their consumption of reactiveenergy.The installation of power-factor correctingcapacitors on installations permits theconsumer to reduce his electricity bill bymaintaining the level of reactive-powerconsumption below a value contractuallyagreed with the power-supply authority.In this particular tariff, reactive energy is billedaccording to the tan ϕ criterion. As previouslynoted:

tan ϕ = Q (kvarh) P (kWh)At the supply service position, the powersupply distributor delivers reactive energyfree, until:c the point at which it reaches 40% of theactive energy (tan ϕ = 0.4) for a maximumperiod of 16 hours each day (from 06-00 h to22-00 h) during the most-heavily loadedperiod (often in winter);c without limitation during light-load periods inwinter, and in spring and summer.During the periods of limitation, reactive-energy consumption exceeding 40% of theactive energy (i.e. tan ϕ > 0.4) is billedmonthly at the current rates.

Thus, the quantity of reactive energy billed inthese periods will be:kvarh (to be billed) = kWh (tan ϕ - 0.4) wherekWh is the active energy consumed duringthe periods of limitation, and kWh tan ϕ is thetotal reactive energy during a period oflimitation, and 0.4 kWh is the amount ofreactive energy delivered free during a periodof limitation.tan ϕ = 0.4 corresponds to a PF of 0.93 sothat, if steps are taken to ensure that duringthe limitation periods the PF never falls below0.93, the consumer will have nothing to payfor the reactive power consumed.Against the financial advantages of reducedbilling, the consumer must balance the cost ofpurchasing, installing and maintaining thepower-factor-improvement capacitors andcontrolling switchgear, automatic controlequipment (where stepped levels ofcompensation are required) together with theadditional kWh consumed by the dielectriclosses of the capacitors, etc.It may be found that it is more economic toprovide partial compensation only, and thatpaying for some of the reactive energyconsumed is less expensive than providing100% compensation.The question of power-factor correction is amatter of optimization, except in very simplecases.

2.2 technical/economic optimization

power factor improvement allows theuse of smaller transformers,switchgear and cables, etc. as wellas reducing power losses andvoltage drop in an installation.

A high power factor allows the optimization ofthe components of an installation.Overating of certain equipment can beavoided, but to achieve the best results, thecorrection should be effected as close to theindividual items of inductive plant as possible.

reduction of cable sizeTable E8 shows the required increase in thesize of cables as the power factor is reducedfrom unity to 0.4.multiplying factor 1 1.25 1.67 2.5for the cross-sectionalarea of the cable core(s)cos ϕ 1 0.8 0.6 0.4

table E8: multiplying factor for cable sizeas a function of cos ϕ.

reduction of losses (P, kW)in cablesLosses in cables are proportional to thecurrent squared, and are measured by thekWh meter for the installation. Reduction ofthe total current in a conductor by 10% forexample, will reduce the losses by almost20%.

reduction of voltage dropPF correction capacitors reduce or evencancel completely the (inductive) reactivecurrent in upstream conductors, therebyreducing or eliminating voltage drops.Note: Overcompensation will produce avoltage rise at the capacitors.

increase in available powerBy improving the power factor of a loadsupplied from a transformer, the currentthrough the transformer will be reduced,thereby allowing more load to be added.In practice, it may be less expensive toimprove the power factor*, than to replace thetransformer by a larger unit.This matter is further elaborated in clause 6.

* Since other benefits accrue from a highvalue of PF, as previously noted.


E

3. how to improve the power factor

3.1 theoretical principles

improving the power factor of aninstallation requires a bank ofcapacitors which acts as a source ofreactive energy. This arrangement issaid to provide reactive energycompensation.

Example:A motor consumes 100 kW at a PF of 0.75(i.e. tan ϕ = 0.88).To improve the PF to 0.93 (i.e. tan ϕ = 0.4),the reactive power of the capacitor bank mustbe : Qc = 100 (0.88 - 0.4) = 48 kvarThe selected level of compensation and thecalculation of rating for the capacitor bankdepend on the particular installation. Thefactors requiring attention are explained in ageneral way in clause 5, and in clauses 6 and7 for transformers and motors.

Note:Before embarking on a compensation project,a number of precautions should be observed.In particular, oversizing of motors should beavoided, as well as the operation of themotors in an unloaded condition. In this lattercondition, the reactive energy consumed bya motor results in a very low power factor(≈ 0.17); this is because the kW taken by themotor (when it is unloaded) are very small.

Qc

ϕ ϕ'

P

S

S'

Q

Q'

fig. E10: diagram showing the principle ofcompensation: Qc = P (tan ϕ - tan ϕ').

Figure E10 uses the power diagramdiscussed in sub-clause 1.3 (figure E3) toillustrate the principle of compensation byreducing a large reactive power Q to asmaller value Q' by means of a bank ofcapacitors having a reactive power Qc.In doing so, the magnitude of the apparentpower S is seen to reduce to S'.

An inductive load having a low power factorrequires the generators andtransmission/distribution systems to passreactive current (lagging the system voltageby 90 degrees) with associated power lossesand exaggerated voltage drops, as noted insub-clause 1.1.If a bank of shunt capacitors is added to theload, its (capacitive) reactive current will takethe same path through the power system asthat of the load reactive current.Since, as pointed out in sub-clause 1.1, thiscapacitive current Ic (which leads the systemvoltage by 90 degrees) is in direct phaseopposition to the load reactive current (IL), thetwo components flowing through the samepath will cancel each other, such thatif the capacitor bank is sufficiently large andIc = IL there will be no reactive current flow inthe system upstream of the capacitors.This is indicated in figure E9 (a) and (b) whichshow the flow of the reactive components ofcurrent only.In this figure:R represents the active-power elements ofthe loadL represents the (inductive) reactive-powerelements of the loadC represents the (capacitive) reactive-powerelements of the power-factor correctionequipment (i.e. capacitors).

IC IL

C L RILIL - IC

loada) reactive current components only flow pattern

IC IL

C L RIL = ICIL - IC = 0

loadb) when IC = IL, all reactive power is supplied from the capacitor bank

IC IL

C L RIR + ILIR IR

loadc) with load current added to case (b)

fig. E9: showing the essential features ofpower-factor correction.

It will be seen from diagram (b) of figure E9,that the capacitor bank C appears to besupplying all the reactive current of the load.For this reason, capacitors are sometimesreferred to as "generators of lagging vars".In diagram (c) of figure E9, the active-powercurrent component has been added, andshows that the (fully-compensated) loadappears to the power system as having apower factor of 1.In general, it is not economical to fullycompensate an installation.


E

3.2 by using what equipment?

compensation at L.V.At low voltage, compensation is providedby:c fixed-valued capacitor,c equipment providing automatic regulation,or banks which allow continuous adjustmentaccording to requirements, as loading of theinstallation changes.

Note:When the installed reactive power ofcompensation exceeds 800 kvar, and theload is continuous and stable, it is often foundto be economically advantageous to instalcapacitor banks at high voltage.

compensation can be carried out bya fixed value of capacitance infavourable circumstances.

fixed capacitors

This arrangement employs one or morecapacitor(s) to form a constant level ofcompensation. Control may be:c manual: by circuit breaker or load-breakswitch,c semi-automatic: by contactor,c direct connection to an appliance andswitched with it.These capacitors are applied:c at the terminals of inductive devices(motors and transformers),c at busbars supplying numerous smallmotors and inductive appliance for whichindividual compensation would be too costly,c in cases where the level of load isreasonably constant.

fig. E11: example of fixed-value-compensation capacitors.

automatic capacitor banks

This kind of equipment provides automaticcontrol of compensation, maintaining withinclose limits, a selected level of power factor.Such equipment is applied at points in aninstallation where the active-power and/orreactive-power variations are relatively large,for example:c at the busbars of a general powerdistribution board,c at the terminals of a heavily-loaded feedercable.

compensation is more-commonlyeffected by means of anautomatically-controlled steppedbank of capacitors.

fig. E12: example ofautomatic-compensation-regulatingequipment.


E

3. how to improve the power factor (continued)

3.2 by using what equipment? (continued)

automatically-regulated banks ofcapacitors allow an immediateadaptation of compensation to matchthe level of load.

the principles of, and reasons,

for using automatic

compensation

A bank of capacitors is divided into a numberof sections, each of which is controlled by acontactor. Closure of a contactor switches itssection into parallel operation with othersections already in service. The size of thebank can therefore be increased ordecreased in steps, by the closure andopening of the controlling contactors.A control relay monitors the power factor ofthe controlled circuit(s) and is arranged toclose and open appropriate contactors tomaintain a reasonably constant systempower factor (within the tolerance imposed bythe size of each step of compensation). Thecurrent transformer for the monitoring relay

must evidently be placed on one phase of theincoming cable which supplies the circuit(s)being controlled, as shown in figure E13.By closely matching compensation to thatrequired by the load, the possibility ofproducing overvoltages at times of low loadwill be avoided, thereby preventing anovervoltage condition, and possible damageto appliances and equipment.Overvoltages due to excessive reactivecompensation depend partly on the value ofsource impedance.

varmetricrelay

CT In / 5 A cl 1

fig. E13: the principle of automatic-compensation control.

3.3 the choice between a fixed or automatically-regulated bank of capacitors

commonly-applied rulesWhere the kvar rating of the capacitors is lessthan, or equal to 15% of the supply-transformer rating, a fixed value ofcompensation is appropriate.Above the 15% level, it is advisable to installan automatically-controlled bank ofcapacitors.

The location of low-voltage capacitors in aninstallation constitutes the mode ofcompensation, which may be global (onelocation for the entire installation), partial(section-by-section), local (at each individualdevice), or some combination of the lattertwo. In principle, the ideal compensation isapplied at a point of consumption and at thelevel required at any instant.In practice, technical and economic factorsgovern the choice.


E

4. where to install correction capacitors

4.1 global compensation

where a load is continuous andstable, global compensation can beapplied.

principle

The capacitor bank is connected to thebusbars of the main LV distribution board forthe installation, and remains in service duringthe period of normal load.

advantagesThe global type of compensation:c reduces the tariff penalties for excessiveconsumption of kvars,c reduces the apparent power kVA demand,on which standing charges are usually based,c relieves the supply transformer, which isthen able to accept more load if necessary.

commentsc reactive current still flows in all conductorsof cables leaving (i.e. downstream of) themain LV distribution board,c for the above reason, the sizing of thesecables, and power losses in them, are notimproved by the global mode ofcompensation. fig. E14: global compensation.

4.2 compensation by sector

compensation by sector isrecommended when the installationis extensive, and where the load/timepatterns differ from one part of theinstallation to another.

principleCapacitor banks are connected to busbars ofeach local distribution board, as shown infigure E14.A significant part of the installation benefitsfrom this arrangement, notably the feedercables from the main distribution board toeach of the local distribution boards at whichthe compensation measures are applied.

advantagesThe compensation by sector:c reduces the tariff penalties for excessiveconsumption of kvars,c reduces the apparent power kva demand,on which standing charges are usually based,c relieves the supply transformer, which isthen able to accept more load if necessary,c the size of the cables supplying the localdistribution boards may be reduced, or willhave additional capacity for possible loadincreases,c losses in the same cables will be reduced.

comments

c reactive current still flows in all cablesdownstream of the local distribution boards,c for the above reason, the sizing of thesecables, and the power losses in them, are notimproved by compensation by sector,c where large changes in loads occur, thereis always a risk of overcompensation andconsequent overvoltage problems.

fig. E15: compensation by sector.

M M M M

n°1

M M M M

n°2 n°2

n°1

kWkvar

conductors


E

4. where to install correction capacitors (continued)

individual compensation should beconsidered when the power of motoris significant with respect to power ofthe installation.

4.3 individual compensation

principle

Capacitors are connected directly to theterminals of inductive plant (notably motors,see further in Clause 7). Individualcompensation should be considered whenthe power of the motor is significant withrespect to the declared power requirement(kVA) of the installation.The kvar rating of the capacitor bank is in theorder of 25% of the kW rating of the motor.Complementary compensation at the origin ofthe installation (transformer) may also bebeneficial.

advantagesIndividual compensation:c reduces the tariff penalties for excessiveconsumption of kvars,c reduces the apparent power kVA demand,c reduces the size of all cables as well as thecable losses.

commentsc significant reactive currents no longer existin the installation.

M M M M

n°1

n°2n°2

n°3 n°3n°3 n°3

fig. E16: individual compensation.


E

5. how to decide the optimum level of compensation

5.1 general method

listing of reactive powerdemands at the design stage

This listing can be made in the same way(and at the same time) as that for the powerloading described in chapter B.The levels of active and reactive powerloading, at each level of the installation(generally at points of distribution andsub-distribution of circuits) can then bedetermined.

5.2 simplified method

general principleAn approximate calculation is generallyadequate for most practical cases, and maybe based on the assumption of a powerfactor of 0.8 (lagging) before compensation.In order to improve the power factor to avalue sufficient to avoid tariff penalties (thisdepends on local tariff structures, but isassumed here to be 0.93) and to reducelosses, volt-drops, etc. in the installation,reference can be made to table E17.From the table, it can be seen that, to raisethe power factor of the installation from 0.8to 0.93 will require 0.355 kvar per kW of load.The rating of a bank of capacitors at thebusbars of the main distribution boardof the installation would beQ (kvar) = 0.355 x P (kW).This simple approach allows a rapiddetermination of the compensation capacitorsrequired, albeit in the global, partial orindependent mode.ExampleIt is required to improve the power factor of a666 kVA installation from 0.75 to 0.928.The active power demand is666 x 0.75 = 500 kW.In table E17, the intersection of the rowcos ϕ = 0.75 (before correction) with thecolumn cos ϕ = 0.93 (after correction)indicates a value of 0.487 kvar ofcompensation per kW of load.For a load of 500 kW, therefore,500 x 0.487 = 244 kvar of capacitivecompensation is required.

Note: this method is valid for any voltagelevel, i.e. is independent of voltage.

technical-economic optimizationfor an existing installation

The optimum rating of compensationcapacitors for an existing installation can bedetermined from the following principalconsiderations:c electricity bills prior to the installation ofcapacitors,c future electricity bills anticipated followingthe installation of capacitors,c costs of:v purchase of capacitors and controlequipment (contactors, relaying, cabinets,etc.),v installation and maintenance costs,v cost of dielectric heating losses in thecapacitors, versus reduced losses in cables,transformer, etc., following the installation ofcapacitors.Several simplified methods applied to typicaltariffs (common in Europe) are shown insub-clauses 5.3 and 5.4.


E

5. how to decide the optimum level of compensation (continued)

before kvar rating of capacitor bank to install per kW of load, to improve cos ϕ (the power factor) or tan ϕ,compensation to a given value

tan ϕ 0,75 0.59 0.48 0.46 0.43 0.40 0.36 0.33 0.29 0.25 0.20 0.14 0.0tan ϕ cos ϕ cos ϕ 0.80 0.86 0.90 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1

2.29 0.40 1.557 1.691 1.805 1.832 1.861 1.895 1.924 1.959 1.998 2.037 2.085 2.146 2.288

2.22 0.41 1.474 1.625 1.742 1.769 1.798 1.831 1.840 1.896 1.935 1.973 2.021 2.082 2.225

2.16 0.42 1.413 1.561 1.681 1.709 1.738 1.771 1.800 1.836 1.874 1.913 1.961 2.022 2.1642.10 0.43 1.356 1.499 1.624 1.651 1.680 1.713 1.742 1.778 1.816 1.855 1.903 1.964 2.107

2.04 0.44 1.290 1.441 1.558 1.585 1.614 1.647 1.677 1.712 1.751 1.790 1.837 1.899 2.041

1.98 0.45 1.230 1.384 1.501 1.532 1.561 1.592 1.628 1.659 1.695 1.737 1.784 1.846 1.9881.93 0.46 1.179 1.330 1.446 1.473 1.502 1.533 1.567 1.600 1.636 1.677 1.725 1.786 1.929

1.88 0.47 1.130 1.278 1.397 1.425 1.454 1.485 1.519 1.532 1.588 1.629 1.677 1.758 1.881

1.83 0.48 1.076 1.228 1.343 1.370 1.400 1.430 1.464 1.497 1.534 1.575 1.623 1.684 1.8261.78 0.49 1.030 1.179 1.297 1.326 1.355 1.386 1.420 1.453 1.489 1.530 1.578 1.639 1.782

1.73 0.50 0.982 1.232 1.248 1.276 1.303 1.337 1.369 1.403 1.441 1.481 1.529 1.590 1.732

1.69 0.51 0.936 1.087 1.202 1.230 1.257 1.291 1.323 1.357 1.395 1.435 1.483 1.544 1.686

1.64 0.52 0.894 1.043 1.160 1.188 1.215 1.249 1.281 1.315 1.353 1.393 1.441 1.502 1.6441.60 0.53 0.850 1.000 1.116 1.144 1.171 1.205 1.237 1.271 1.309 1.349 1.397 1.458 1.600

1.56 0.54 0.809 0.959 1.075 1.103 1.130 1.164 1.196 1.230 1.268 1.308 1.356 1.417 1.559

1.52 0.55 0.769 0.918 1.035 1.063 1.090 1.124 1.156 1.190 1.228 1.268 1.316 1.377 1.5191.48 0.56 0.730 0.879 0.996 1.024 1.051 1.085 1.117 1.151 1.189 1.229 1.277 1.338 1.480

1.44 0.57 0.692 0.841 0.958 0.986 1.013 1.047 1.079 1.113 1.151 1.191 1.239 1.300 1.442

1.40 0.58 0.665 0.805 0.921 0.949 0.976 1.010 1.042 1.076 1.114 1.154 1.202 1.263 1.4051.37 0.59 0.618 0.768 0.884 0.912 0.939 0.973 1.005 1.039 1.077 1.117 1.165 1.226 1.368

1.33 0.60 0.584 0.733 0.849 0.878 0.905 0.939 0.971 1.005 1.043 1.083 1.131 1.192 1.334

1.30 0.61 0.549 0.699 0.815 0.843 0.870 0.904 0.936 0.970 1.008 1.048 1.096 1.157 1.299

1.27 0.62 0.515 0.665 0.781 0.809 0.836 0.870 0.902 0.936 0.974 1.014 1.062 1.123 1.2651.23 0.63 0.483 0.633 0.749 0.777 0.804 0.838 0.870 0.904 0.942 0.982 1.030 1.091 1.233

1.20 0.64 0.450 0.601 0.716 0.744 0.771 0.805 0.837 0.871 0.909 0.949 0.997 1.058 1.200

1.17 0.65 0.419 0.569 0.685 0.713 0.740 0.774 0.806 0.840 0.878 0.918 0.966 1.007 1.1691.14 0.66 0.388 0.538 0.654 0.682 0.709 0.743 0.775 0.809 0.847 0.887 0.935 0.996 1.138

1.11 0.67 0.358 0.508 0.624 0.652 0.679 0.713 0.745 0.779 0.817 0.857 0.905 0.966 1.108

1.08 0.68 0.329 0.478 0.595 0.623 0.650 0.684 0.716 0.750 0.788 0.828 0.876 0.937 1.079

1.05 0.69 0.299 0.449 0.565 0.593 0.620 0.654 0.686 0.720 0.758 0.798 0.840 0.907 1.0491.02 0.70 0.270 0.420 0.536 0.564 0.591 0.625 0.657 0.691 0.729 0.769 0.811 0.878 1.020

0.99 0.71 0.242 0.392 0.508 0.536 0.563 0.597 0.629 0.663 0.701 0.741 0.783 0.850 0.992

0.96 0.72 0.213 0.364 0.479 0.507 0.534 0.568 0.600 0.634 0.672 0.712 0.754 0.821 0.9630.94 0.73 0.186 0.336 0.452 0.480 0.507 0.541 0.573 0.607 0.645 0.685 0.727 0.794 0.936

0.91 0.74 0.159 0.309 0.425 0.453 0.480 0.514 0.546 0.580 0.618 0.658 0.700 0.767 0.909

0.88 0.75 0.132 0.82 0.398 0.426 0.453 0.487 0.519 0.553 0.591 0.631 0.673 0.740 0.8820.86 0.76 0.105 0.255 0.371 0.399 0.426 0.460 0.492 0.526 0.564 0.604 0.652 0.713 0.855

0.83 0.77 0.079 0.229 0.345 0.373 0.400 0.434 0.466 0.500 0.538 0.578 0.620 0.687 0.829

0.80 0.78 0.053 0.202 0.319 0.347 0.374 0.408 0.440 0.474 0.512 0.552 0.594 0.661 0.803

0.78 0.79 0.026 0.176 0.292 0.320 0.347 0.381 0.413 0.447 0.485 0.525 0.567 0.634 0.7760.75 0.80 0.150 0.266 0.294 0.321 0.355 0.387 0.421 0.459 0.499 0.541 0.608 0.750

0.72 0.81 0.124 0.240 0.268 0.295 0.329 0.361 0.395 0.433 0.473 0.515 0.582 0.724

0.70 0.82 0.098 0.214 0.242 0.269 0.303 0.335 0.369 0.407 0.447 0.489 0.556 0.6980.67 0.83 0.072 0.188 0.216 0.243 0.277 0.309 0.343 0.381 0.421 0.463 0.530 0.672

0.65 0.84 0.046 0.162 0.190 0.217 0.251 0.283 0.317 0.355 0.395 0.437 0.504 0.645

0.62 0.85 0.020 0.136 0.164 0.191 0.225 0.257 0.291 0.329 0.369 0.417 0.478 0.6200.59 0.86 0.109 0.140 0.167 0.198 0.230 0.264 0.301 0.343 0.390 0.450 0.593

0.57 0.87 0.083 0.114 0.141 0.172 0.204 0.238 0.275 0.317 0.364 0.424 0.567

0.54 0.88 0.054 0.085 0.112 0.143 0.175 0.209 0.246 0.288 0.335 0.395 0.538

0.51 0.89 0.028 0.059 0.086 0.117 0.149 0.183 0.230 0.262 0.309 0.369 0.5120.48 0.90 0.031 0.058 0.089 0.121 0.155 0.192 0.234 0.281 0.341 0.484

table E17: kvar to be installed per kW of load, to improve the power factor of an installation.


E

Qc

ϕ ϕ'

P = 85.4 kW

S

S'

Q

Q'

cos ϕ = 0.7cos ϕ' = 0.95

S = 122 kVAS' = 90 kVAQ = 87.1 kvarQc = 59 kvarQ' = 28.1 kvar

5.3 method based on the avoidance of tariff penalties

in the case of certain (common)types of tariff, an examination ofseveral bills covering the mostheavily-loaded period of the yearallows determination of the kvar levelof compensation required to avoidkvarh (reactive-energy) charges.

the pay-back period of a bankof power-factor-improvementcapacitors and associated equipmentis generally about 18 months.

The following method allows calculation ofthe rating of a proposed capacitor bank,based on billing details, where the tariffstructure corresponds with (or is similar to)the one described in sub-clause 2.1 of thischapter.The method determines the minimumcompensation required to avoid thesecharges which are based on kvarhconsumption.The procedure is as follows:c refer to the bills covering consumption forthe 5 months of winter (in France these areNovember to March inclusive).Note: in tropical climates the summer monthsmay constitute the period of heaviest loadingand highest peaks (owing to extensive air-conditioning loads) so that a consequentvariation of high-tariff periods is necessary inthis case. The remainder of this example willassume Winter conditions in France.c identify the line on the bills referring to"reactive-energy consumed" and "kvarh to becharged". Choose the bill which shows thehighest charge for kvarh (after checking thatthis was not due to some exceptionalsituation).For example: 15,966 kvarh in January.c evaluate the total period of loadedoperation of the installation for that month, forinstance: 220 hours (22 days x 10 hours).The hours which must be counted are thoseoccurring during the heaviest load and thehighest peak loads occurring on the powersystem. These are given in the tariffdocuments, and are (commonly) during a16 hour period each day, either from 06.00 Hto 22.00 H or from 07.00 H to 23.00 Haccording to the region. Outside theseperiods, no charge is made for kvarhconsumption.

c the necessary value of compensation

in kvar = kvarh billed = Qc number of hours of operation*

* in the billing period, during the hours forwhich reactive energy is chargedfor the case considered above:

Qc =15,966 kvarh = 73 kvar 220 hThe rating of the installed capacitor bank isgenerally chosen to be slightly larger thanthat calculated.Certain manufacturers can provide "slide-rules" especially designed to facilitate thesekinds of calculation, according to particulartariffs. These devices and accompanyingdocumentation advise on suitable equipmentand control schemes, as well as drawingattention to constraints imposed by harmonicvoltages on the power system. Such voltagesrequire either overdimensioned capacitors (interms of heat-dissipation, voltage and currentratings) and/or harmonic-suppressioninductors or filters (see Appendix E3).

5.4 method based on reduction of declared maximum apparent power (kVA)

for 2-part tariffs based partly on adeclared value of kVA, a table (E16)allows determination of the kvar ofcompensation required to reduce thevalue of kVA declared, and to avoidexceeding it.

For consumers whose tariffs are based on afixed charge per kVA declared, plus a chargeper kWh consumed, it is evident that areduction in declared kVA would be beneficial.The diagram of figure E18 shows that as thepower factor improves, the kVA valuediminishes for a given value of kW (P).The improvement of the power factor isaimed at (apart from other advantagespreviously mentioned) reducing the declaredlevel and never exceeding it, therebyavoiding the payment of an excessive priceper kVA during the periods of excess, and/ortripping of the the main circuit breaker.Table E17 indicates the value of kvar ofcompensation per kW of load, required toimprove from one value of power factor toanother.

Example:A supermarket has a declared load of122 kVA at a power factor of 0.7 lagging, i.e.an active-power load of 85.4 kW.The particular contract for this consumer wasbased on stepped values of declared kVA(in steps of 6 kVA up to 108 kVA, and 12 kVAsteps above that value, this is a commonfeature in many types of two-part tariff).In the case being considered, the consumerwas billed on the basis of 132 kVA.Referring to table E17, it can be seen that a60 kvar bank of capacitors will improve thepower factor of the load from 0.7 to 0.95(0.691 x 85.4 = 59 kvar in the table).The declared value of kVA will then be85.4 = 90 kVA, i.e. an improvement of 30%.0.95

fig. E18: reduction of declared maximumkVA by power-factor improvement.


E

6. compensation at the terminals of a transformer

6.1 compensation to increase the available active power output

the installation of a bank ofcapacitors can avoid the need tochange a transformer in the event ofa load increase.

Steps similar to those taken to reduce thedeclared maximum kVA, i.e. improvement ofthe load power factor, as discussed in sub-clause 5.4, will maximise the availabletransformer capacity, i.e. to supply moreactive power.Cases can arise where the replacement of atransformer by a larger unit, to cater for loadgrowth, may be avoided by this means.Table E20 shows directly the power (kW)capability of fully-loaded transformers atdifferent load power factors, from which theincrease of active-power output, as the valueof power factor increases, can be obtained.

Example:Refer to figure E19.An installation is supplied from a 630 kVAtransformer loaded at 450 kW (P1) with amean power factor of 0.8 lagging.

The apparent power S1 = 450 = 562 kVA 0.8

The corresponding reactive power

The anticipated load increase P2 = 100 kW ata power factor of 0.7 lagging.

The apparent power S2 = 100 = 143 kVA 0.7

The corresponding reactive power

What is the minimum value of capacitive kvarto be installed, in order to avoid a change oftransformer?Total power now to be supplied:P = P1 + P2 = 550 kW.The maximum reactive power capability ofthe 630 kVA transformer when delivering550 kW is:

Total reactive power required by theinstallation before compensation:Q1 + Q2 = 337 + 102 = 439 kvar.So that the minimum size of capacitor bank toinstal:Qkvar = 439 - 307 = 132 kvar.It should be noted that this calculation has nottaken account of peak loads and theirduration.The best possible improvement, i.e.correction which attains a PF of 1 wouldpermit a power reserve for the transformer of630 - 550 = 80 kW. The capacitor bank wouldthen have to be rated at 439 kvar.

Q1= 337 kvar12

12S P− =

Q2 = 1 kvar22

22 02S P− =

Qm = S -P2 2

Qm = 6 - 550 = 307 kvar2 230

Q

Q

PP1

S1S

Q1

P2

S2 Q2

Qm

fig. E19: compensation Q allows theinstallation-load extension S2 to beadded, without the need to replace theexisting transformer, the output of whichis limited to S.

tan ϕ cos ϕ nominal kVA rating of transformers (in kVA)100 160 250 315 400 500 630 800 1000 1250 1600 2000

0.00 1 100 160 250 315 400 500 630 800 1000 1250 1600 20000.20 0.98 98 157 245 309 392 490 617 784 980 1225 1568 19600.29 0.96 96 154 240 302 384 480 605 768 960 1200 1536 19200.36 0.94 94 150 235 296 376 470 592 752 940 1175 1504 18800.43 0.92 92 147 230 290 368 460 580 736 920 1150 1472 18400.48 0.90 90 144 225 284 360 450 567 720 900 1125 1440 18000.54 0.88 88 141 220 277 352 440 554 704 880 1100 1408 17600.59 0.86 86 138 215 271 344 430 541 688 860 1075 1376 17200.65 0.84 84 134 210 265 336 420 529 672 840 1050 1344 16800.70 0.82 82 131 205 258 328 410 517 656 820 1025 1312 16400.75 0.80 80 128 200 252 320 400 504 640 800 1000 1280 16000.80 0.78 78 125 195 246 312 390 491 624 780 975 1248 15600.86 0.76 76 122 190 239 304 380 479 608 760 950 1216 15200.91 0.74 74 118 185 233 296 370 466 592 740 925 1184 14800.96 0.72 72 115 180 227 288 360 454 576 720 900 1152 14401.02 0.70 70 112 175 220 280 350 441 560 700 875 1120 1400

table E20: active-power capability of fully-loaded transformers, when supplying loads at different values of power factor.


E

XL

Esource

Vload

6.2 compensation of reactive energy absorbed by the transformer

where metering is carried out at theHV side of a transformer, thereactive-energy losses in thetransformer may (depending on thetariff) need to be compensated.

the nature of transformer

inductive reactances

All previous references have been to shunt-connected devices such as those used innormal loads, and PF-correcting capacitorbanks etc. The reason for this is that shunt-connected plant requires (by far) the largestquantities of reactive energy in powersystems; however, series-connectedreactances, such as the inductive reactancesof power lines and the leakage reactance oftransformer windings, etc., also absorbreactive energy.Where metering is carried out at the HV sideof a transformer, the reactive-energy losses inthe transformer may (depending on the tariff)need to be compensated.As far as reactive-energy losses only areconcerned, a transformer may be

represented by the elementary diagram offigure E22. All reactance values are referredto the secondary side of the transformer,where the shunt branch represents themagnetizing-current path. The magnetizingcurrent remains practically constant (at about1.8 % of full-load current) from no load to fullload, in normal circumstances, i.e. with aconstant primary voltage, so that a shuntcapacitor of fixed value can be installed at theHV or LV side, to compensate for the reactiveenergy absorbed.

fig. E22: transformer reactances(per phase).

the reactive power absorbed by atransformer cannot be neglected,and can amount to (about) 5% of thetransformer rating when supplying itsfull load. Compensation can beprovided by a bank of capacitors.

in transformers, reactive power isabsorbed by both shunt(magnetizing) and series (leakageflux) reactances. Completecompensation can be providedby a bank of shunt-connected LVcapacitors.

reactive-power absorption

in series-connected

(leakage flux) reactance

A simple illustration of this phenomenon isgiven by the vector diagram of figure E21.The reactive-current component through theload = I sin ϕ so that kvarL = VI sin ϕ.The reactive-current component from thesource = I sin ϕ' so thatkvars = EI sin ϕ'where V and E are expressed in kV.It can be seen that E > V and sin ϕ' > sin ϕ.The difference between EI sin ϕ' and VI sin ϕgives the kvar per phase absorbed by XL.It can be shown that this kvar value is equalto I2XL (which is analogous to the I2R active-power (kW) losses due to the seriesresistance of power lines, etc.).From the I2XL formula it is very simple todeduce the kvar absorbed at any load valuefor a given transformer, as follows:If per-unit values are used (instead ofpercentage values) direct multiplicationof I and XL can be carried out.

fig. E21: reactive power absorption byseries inductance.

Example:A 630 kVA transformer with a short-circuitreactance voltage of 4% is fully loaded.What is its reactive-power (kvar) loss?4% = 0.04 pu Ipu = 1loss = I2XL = 12 x 0.04 = 0.04 pu kvarwhere 1 pu = 630 kVA

The 3-phase kvar lossesare 630 x 0.04 = 25.2 kvar(or, quite simply, 4% of 630 kVA).At half load i.e. I = 0.5 pu the losseswill be 0.52 x 0.04 = 0.01 pu =630 x 0.01 = 6.3 kvar and so on...This example, and the vector diagram offigure E21 show that:c the power factor at the primary side of aloaded transformer is different (normallylower) than that at the secondary side (due tothe absorption of vars),c that full-load kvar losses due to leakagereactance are equal to the transformerpercentage reactance (4% reactance meansa kvar loss equal to 4% of the kVA rating ofthe transformer),c that kvar losses due to leakage reactancevary according to the current (or kVA loading)squared.To determine the total kvar losses of atransformer the constant magnetizing-currentcircuit losses (approx. 1.8% of thetransformer kVA rating) must be added to theforegoing "series" losses.Table E24 shows the no-load and full-loadkvar losses for typical distributiontransformers.In principle, series inductances can becompensated by fixed series capacitors (as iscommonly the case for long HV transmissionlines). This arrangement is operationallydifficult, however, so that, at the voltage levelscovered by this guide, shunt compensation isalways applied.In the case of HV metering, it is sufficient toraise the power factor to a point where thetransformer plus load reactive-powerconsumption is below the level at which abilling charge is made. This level depends onthe tariff, but often corresponds to a tan ϕvalue of 0.31 (cos ϕ of 0.955).As a matter of interest, the kvar losses in atransformer can be completely compensatedby adjusting the capacitor bank to give theload a (slightly) leading power factor. In sucha case, all of the kvar of the transformer isbeing supplied from the capacitor bank, whilethe input to the HV side of the transformer isat unity power factor, as shown in figure E23.

perfect transformer

primarywinding

secondarywinding

magnetizingreactance

leakage reactance

ϕϕ'

I sin ϕ

I sin ϕ'

V

E

IXL


E

6. compensation at the terminals of a transformer (continued)

6.2 compensation of reactive energy absorbed by the transformer (continued)

ϕ

IXLI

V (load voltage)

E (input voltage)

load current

I0 compensation current

fig. E23: overcompensation of load tocompletely compensate transformerreactive-power losses.

In practical terms, therefore, compensationfor transformer-absorbed kvar is included inthe capacitors primarily intended for power-factor correction of the load, either globally,partially, or in the individual mode.Unlike most other kvar-absorbing items, thetransformer absorption (i.e. the part due tothe leakage reactance) changes significantlywith variations of load level, so that, ifindividual compensation is applied to thetransformer, then an average level of loadingwill have to be assumed.Fortunately, this kvar consumption generallyforms only a relatively small part of the totalreactive power of an installation, and somismatching of compensation at times ofload change is not likely to be a problem.Table E24 indicates typical kvar loss valuesfor the magnetizing circuit (“no-load kvar”columns), as well as for the total losses at fullload, for a standard range of distributiontransformers supplied at 20 kV (which includethe losses due to the leakage reactance).

rated power reactive power (kvar) to be compensatedkVA oil immersed type cast resin type

no load full load no load full load50 1.5 2.9100 2.5 5.9 2.5 8.2160 3.7 9.6 3.7 12.9250 5.3 14.7 5.0 19.5315 6.3 18.3 5.7 24.0400 7.6 22.9 6.0 29.4500 9.5 28.7 7.5 36.8630 11.3 35.7 8.2 45.2800 20 66.8 10.4 57.51 000 24.0 82.6 12.0 71.01 250 27.5 100.8 15.0 88.81 600 32.0 125.9 19.2 113.92 000 38.0 155.3 22.0 140.62 500 45.0 191.5 30.0 178.2

table E24: reactive power consumption of distribution transformers with 20 kVprimary windings.

Note: for a 630 kVA transformer, the range ofkvar losses extends from 11.3 at no load to35.7 kvar at full load. These valuescorrespond closely to those given in theworked example above.


E

7. compensation at the terminals of an induction motor

7.1 connection of a capacitor bank and protection settings

individual motor compensation isrecommended where the motorpower (kVA) is large with respect tothe declared power of the installation.

general precautions

Because of the small kW consumption, thepower factor of a motor is very low at no-loador on light load. The reactive current of themotor remains practically constant at allloads, so that a number of unloaded motorsconstitute a consumption of reactive powerwhich is generally detrimental to aninstallation, for reasons explained inpreceding sections.Two good general rules therefore are thatunloaded motors should be switched off, andmotors should not be oversized (since theywill then be lightly loaded).

connectionThe bank of capacitors should be connecteddirectly to the terminals of the motor.

special motorsIt is recommended that special motors(stepping, plugging, inching, reversingmotors, etc.) should not be compensated.

effect on protection settingsAfter applying compensation to a motor, thecurrent to the motor-capacitor combinationwill be lower than before, assuming the samemotor-driven load conditions. This is becausea significant part of the reactive component ofthe motor current is being supplied from thecapacitor, as shown in figure E25.Where the overcurrent protection devices ofthe motor are located upstream of the motor-capacitor connection (and this will always bethe case for terminal-connected capacitors),the overcurrent relay settings must bereduced in the ratio:cos ϕ before compensation cos ϕ after compensationFor motors compensated in accordance withthe kvar values indicated in Table E28(maximum values recommended foravoidance of self-excitation of standardinduction motors, as discussed insub-clause 7.2), the above-mentioned ratiowill have a value similar to that indicated forthe corresponding motor speed in Table E26.

beforecompensation

activepower

aftercompensation

reactivepowersuppliedby capacitor

powermadeavailable

C

transformer

motorM M

fig. E25: before compensation, thetransformer supplies all the reactivepower; after compensation, the capacitorsupplies a large part of the reactive power.

speed in R.P.M. reduction factor750 0.881000 0.901500 0.913000 0.93

table E26: reduction factor for overcurrentprotection after compensation.


E

7. compensation at the terminals of an induction motor (continued)

7.2 how self-excitation of an induction motor can be avoided

when a capacitor bank is connectedto the terminals of an inductionmotor, it is important to check that thesize of the bank is less than that atwhich self-excitation can occur.

When a motor is driving a high-inertia load,the motor will continue to rotate (unlessdeliberately braked) after the motor supplyhas been switched off.The "magnetic inertia" of the rotor circuitmeans that an emf will be generated in thestator windings for a short period afterswitching off, and would normally reduce tozero after 1 or 2 cycles, in the case of anuncompensated motor.Compensation capacitors however, constitutea 3-phase "wattless" load for this decayingemf, which causes capacitive currents to flowthrough the stator windings. These statorcurrents will produce a rotating magnetic fieldin the rotor which acts exactly along the sameaxis and in the same direction as that of thedecaying magnetic field.The rotor flux consequently increases; thestator currents increase; and the voltage atthe terminals of the motor increases;sometimes to dangerously-high levels.This phenomenon is known as self-excitationand is one reason why a.c. generators arenot normally operated at leading powerfactors, i.e. there is a tendency tospontaneously (and uncontrollably) self-excite.

Notes1. The characteristics of a motor being drivenby the inertia of the load are not rigorouslyidentical to its no-load characteristics. Thisassumption, however, is sufficiently accuratefor practical purposes.2. With the motor acting as a generator, thecurrents circulating are largely reactive, sothat the braking (retarding) effect on themotor is mainly due only to the loadrepresented by the cooling fan in the motor.3. The (almost 90° lagging) current takenfrom the supply in normal circumstances bythe unloaded motor, and the (almost 90°leading) current supplied to the capacitors bythe motor acting as a generator, both havethe same phase relationship to the terminalvoltage. It is for this reason that the twocharacteristics may be superimposed on thegraph.In order to avoid self-excitation as describedabove, the kvar rating of the capacitor bankmust be limited to the following maximumvalue:Qc i 0,9 Io Un e where Io = the no-loadcurrent of the motor and Un = phase-to-phase nominal voltage of the motor in kV.Table E28 gives appropriate values of Qccorresponding to this criterion.

ExampleA 75 kW, 3 000 Rpm, 400 V, 3-phase motormay have a capacitor bank no larger than17 kvar according to Table E28. The tablevalues are, in general, too small toadequately compensate the motor to the levelof cos ϕ normally required. Additionalcompensation can, however, be applied tothe system, for example an overall bank,installed for global compensation of a numberof smaller appliances.High-inertia motors and/or loadsIn any installation where high-inertia motor-driven loads exist, the circuit breakers orcontactors controlling such motors should, inthe event of total loss of power supply, berapidly tripped.If this precaution is not taken, then self-excitation to very high voltages is likely tooccur, since all other banks of capacitors inthe installation will effectively be in parallelwith those of the high-inertia motors.

fig. E27: connection of the capacitor bankto the motor.

The protection scheme for these motorsshould therefore include an overvoltagetripping relay, together with reverse-powerchecking contacts (the motor will feed powerto the rest of the installation, until the storedinertial energy is dissipated).An undervoltage relay would not be suitablebecause the voltage is not only maintained,but will increase, immediately following theloss of power supply.


EIf the capacitor bank associated with a high-inertia motor is larger than that recommendedin Table E28, then it should be separatelycontrolled by a circuit breaker or contactor,which trips in unison with the main motor-controlling circuit breaker or contactor, asshown in figure E27.Closing of the main contactor is commonlysubject to the capacitor contactor beingpreviously closed.

3-phase motors 230/400 Vnominal kvar to be installedpower

speed of rotation (RPM)kW hp 3000 1500 1000 75022 30 6 8 9 1030 40 7.5 10 11 12.537 50 9 11 12.5 1645 60 11 13 14 1755 75 13 17 18 2175 100 17 22 25 2890 125 20 25 27 30110 150 24 29 33 37132 180 31 36 38 43160 218 35 41 44 52200 274 43 47 53 61250 340 52 57 63 71280 380 57 63 70 79355 482 67 76 86 98400 544 78 82 97 106450 610 87 93 107 117

table E28: maximum kvar of P.F.correction applicable to motor terminalswithout risk of self-excitation.

NoteExact sizing of capacitor unit for a particularmotor is only possible when the "no-loadcurrent" or "no-load magnetising" kvar isknown.


E

cos ϕ = 0.75workshop

kVA

kW kvar

630 kVA

400 V

8. example of an installation before and after power-factorcorrection

installation before P.F. correction

c kvarh are billed heavily above the declaredlevel,c apparent power kVA is significantly greaterthan the kW demand,c the corresponding excess current causeslosses (kWh) which are billed,c the installation must be over-dimensioned.* the arrows denote vector quantities.

kVA=kW+kvar→→→

Characteristics of the installation500 kW cos ϕ = 0.75c transformer is overloadedc the power demand is

S = P = 500 = 665 kVA cos ϕ 0.75S = apparent power

c the current flowing into the installationdownstream of the circuit breaker is

I = P = 960 A eU cos ϕ

c losses in cables are calculated as afunction of the current squared: (960)2

P=I2R

cos ϕ = 0.75c reactive energy is supplied through thetransformer and via the installation wiring,c the transformer, circuit breaker, and cablesmust be over-dimensioned.

installation after P.F. correction

kVA=kW+kvar→→→ c the consumption of kvarh is

v eliminated, orv reduced, according to the cos ϕ required,c the tariff penaltiesv for reactive energy where applicablev for the entire bill in some cases areeliminated,c the fixed charge based on kVA demand isadjusted to be close to the active power kWdemand.

cos ϕ = 0.928workshop

kVA

kW

630 kVA

400 V

kvar

Characteristics of the installation500 kW cos ϕ = 0.928c transformer no longer overloadedc the power-demand is 539 kVAc there is 14% spare-transformer capacityavailable.

c the current flowing into the installationthrough the circuit breaker is 778 A.

c the losses in the cables are

reduced to (778)2 = 65% of the former value, (960)2

thereby economizing in kWh consumed.

cos ϕ = 0.928c reactive energy is supplied by the capacitorbank.

Capacitor bank rating is 250 kvarin 5 automatically-controlled steps of 50 kvar.

Note: In fact, the cos ϕ of the workshopremains at 0.75 but cos ϕ for all theinstallation upstream of the capacitor bankto the transformer LV terminals is 0.928.As mentioned in Sub-clause 6.2, the cos ϕat the HV side of the transformer will beslightly lower, * due to the reactive powerlosses in the transformer.* moreso in the pre-corrected case.

fig. E29: technical-economic comparison of an installation before and after power-factor correction.

*


E

9. the effect of harmonics on the rating of a capacitor bank

9.1 problems arising from power-system harmonics

Equipment which uses power electronicscomponents (variable-speed motorcontrollers, thyristor-controlled rectifiers, etc.)have, in recent years, considerably increasedthe problems caused by harmonics in power-supply systems.Harmonics have existed from the earliestdays of the industry and were (and still are)caused by the non-linear magnetizingimpedances of transformers, reactors,fluorescent lamp ballasts, etc.Harmonics on symmetrical 3-phase powersystems are generally* odd-numbered: 3rd,5th, 7th, 9th..., and the magnitude decreasesas the order of the harmonic increases.

All of these features may be used in variousways to reduce specific harmonics tonegligible values - total elimination is notpossible. In this section, practical means ofreducing the influence of harmonics arerecommended, with particular reference tocapacitor banks.Capacitors are especially sensitive toharmonic components of the supply voltagedue to the fact that capacitive reactancedecreases as the frequency increases. Inpractice, this means that a relatively smallpercentage of harmonic voltage can cause asignificant current to flow in the capacitorcircuit.

The presence of harmonic componentscauses the (normally sinusoidal) wave formof voltage or current to be distorted; the

greater the harmonic content, the greater thedegree of distortion .

If the natural frequency of the capacitor bank/power-system reactance combination is closeto a particular harmonic, then partialresonance will occur, with amplified values ofvoltage and current at the harmonicfrequency concerned.In this particular case, the elevated currentwill cause overheating of the capacitor, withdegradation of the dielectric, which may resultin its eventual failure.Several solutions to these problems areavailable, which aim basically at reducing thedistortion of the supply-voltage wave form,between the equipment causing thedistortion, and the bank of capacitors inquestion. This is generally accomplished byshunt connected harmonic filter and/orharmonic-suppression reactors.

* With the advent of power electronics devices, andassociated non-linear components, even-numberedharmonics are now sometimes encountered.

9.2 possible solutions

harmonics are taken into accountmainly by oversizing capacitors andincluding harmonic-suppressionreactors in series with them.

countering the effects of

harmonics

The presence of harmonics in the supplyvoltage results in abnormally high currentlevels through the capacitors. An allowance ismade for this by designing for an r.m.s. valueof current equal to 1.3 times the nominalrated current.All series elements, such as connections,fuses, switches, etc., associated with thecapacitors are similarly oversized, between1.3 to 1.5 times nominal rating.

Harmonic distortion of the voltage wavefrequently produces a "peaky" wave form, inwhich the peak value of the normal sinusoidalwave is increased. This possibility, togetherwith other overvoltage conditions likely tooccur when countering the effects ofresonance, as described below, are takeninto account by increasing the insulation levelabove that of "standard" capacitors.In many instances, these two countermeasures are all that is necessary to achievesatisfactory operation.

countering the effects

of resonance

Capacitors are linear reactive devices, andconsequently do not generate harmonics.The installation of capacitors in a powersystem (in which the impedances arepredominantly inductive) can, however, resultin total or partial resonance occurring at oneof the harmonic frequencies.The harmonic order ho of the naturalresonant frequency between the systeminductance and the capacitor bank is given by whereSsc = the level of system short-circuit kVA atthe point of connection of the capacitorQ = capacitor bank rating in kvar; and ho =the harmonic order of the natural frequency foi.e. fo/50 for a 50 Hz system, or fo/60 for a60 Hz system.

For example: may give a value forho of 2.93 which shows that the naturalfrequency of the capacitor/system-inductancecombination is close to the 3rd harmonicfrequency of the system.From ho = fo/50 it can be seen thatfo = 50 ho = 50 x 2.93 = 146.5 Hz

The closer a natural frequency approachesone of the harmonics present on the system,the greater will be the (undesirable) effect.In the above example, strong resonantconditions with the 3rd harmonic componentof a distorted wave would certainly occur.S QSC /

S QSC /


E

9. the effect of harmonics on the rating of a capacitor bank (continued)

9.2 possible solutions (continued)

countering the effects

of resonance (continued)In such cases, steps are taken to change thenatural frequency to a value which will notresonate with any of the harmonics known tobe present. This is achieved by the additionof a harmonic-suppression inductorconnected in series with the capacitor bank.On 50 Hz systems, these reactors are oftenadjusted to bring the resonant frequency ofthe combination, i.e. the capacitor bank +reactors to 190 Hz. The reactors are adjustedto 228 Hz for a 60 Hz system.These frequencies correspond to a value forho of 3.8 for a 50 Hz system, i.e.approximately mid-way between the 3rd and5th harmonics.

In this arrangement, the presence of thereactor increases the fundamental-frequency(50 Hz or 60 Hz) current by a small amount(7-8%) and therefore the voltage across thecapacitor in the same proportion.This feature is taken into account, forexample, by using capacitors which aredesigned for 440 V operation on 400 Vsystems.

9.3 choosing the optimum solution

A choice is made from the followingparameters:c Gh = the sum of the kVA ratings of allharmonic-generating devices (staticconverters, inverters, speed controllers, etc.)connected to the busbars from which thecapacitor bank is supplied.If the ratings of some of these devices arequoted in kW only, assume an average powerfactor of 0.7 to obtain the kVA ratings.c Ssc = the 3-phase short-circuit level in kVAat the terminals of the capacitor bank.

c Sn = the sum of the kVA ratings of alltransformers supplying (i.e. directlyconnected to) the system level of which thebusbars form a part.If a number of transformers are operating inparallel, the removal from service of one ormore, will significantly change the values ofSsc and Sn.From these parameters, a choice of capacitorspecification which will ensure an acceptablelevel of operation with the system harmonicvoltages and currents, can be made, byreference to the following table.

capacitors supplied at LV via transformer(s)c general rule valid for any size of transformer

Gh i Ssc Ssc i Gh i Ssc Gh > Ssc 120 120 70 70standard capacitors capacitor voltage rating capacitor voltage rating

increased by 10% increased by 10%(except 230 V units) + harmonic-suppression reactor

c simplified rule if transformer(s) rating Sn i 2 MVAGh i 0.15 Sn 0.15 Sn < Gh i 0.25 Sn 0.25 Sn < Gh i 0.60 Sn Gh > 0.60 Snstandard capacitors capacitor voltage rating capacitor voltage rating

increased by 10% increased by 10% filters(except 230 V units) + harmonic suppression

reactor

table E30: choice of solutions for limiting harmonics associated with a LV capacitor bank.


Eexamples

Three cases are presented, showing(respectively) situations in which standard,overdimensioned, and overdimensioned plusharmonic-suppression-equipped capacitorbanks should be installed.Example 1:500 kVA transformer having 4% short-circuitvoltage.Total rating of harmonic-generating devicesGh = 50 kVA

Ssc = 500 x 100 = 12,500 kVA 4Ssc = 12,500 = 104120 120

Gh = 50 i Ssc 120Solution: use standard capacitors.

Example 2:1,000 kVA transformer having 6% short-circuitvoltage.Total rating of harmonic-generating devicesGh = 220 kVA

Ssc = 1,000 x 100 = 16,667 kVA 6Ssc = 16,667 = 139120 120Ssc = 16,667 = 238 70 70

Gh = 220 is between Ssc and Ssc 120 70Solution: use overated (440 V) capacitors.

Example 3:630 kVA transformer having 4% short-circuitvoltage.Total rating of harmonic-generating devicesGh = 250 kVA

Ssc = 630 x 100 = 15,750 kVA 4Ssc = 15,750 = 225 70 70

Gh = 250 > Scc 70Solution: use overated (440 V) capacitorsand harmonic-suppression reactors.

9.4 possible effects of power-factor-correction capacitors on the power-supply system

it is necessary to ensure thatinteraction between harmonic-generating devices and P.F.correction capacitors, does not resultin unacceptable levels of voltageand/or current wave-form distortionon the power-supply network.

Power-supply authorities generally impose astrict limit on the total-harmonic distortion(THD) permitted at the point of power supplyto a consumer.The degree of distortion is measured as theratio of the r.m.s. value of all harmonicspresent, with respect to the r.m.s. value of thefundamental frequency wave (50 or 60 Hz).For LV loads supplied through a transformerfrom a high-voltage service connection, thismeans that a maximum value of 4 or 5% forvoltage THD is permissible at the LVterminals of the transformer.

If this value of THD is unattainable, thenrecourse has to be made to low-voltage L-Cseries filters. Such filters are shunt-connected, and are tuned to resonate atharmonic frequencies to which they presentpractically zero impedance.Filters connected in this way fortuitously havethe added benefit of contributing to reactive-power compensation for the installation.


E

10. implementation of capacitor banks

10.1 capacitor elements

technology

The capacitors are dry-type units (i.e. are notimpregnated by liquid dielectric) comprisingmetallized polypropylene self-healing film inthe form of a two-film roll.They are protected by a high-quality system(overpressure disconnector used with anHPC fuse) which switches off the capacitor ifan internal fault occurs.The protection scheme operates as follows:c a short-circuit through the dielectric willblow the fuse,c current levels greater than normal, butinsufficient to blow the fuse sometimes occur,e.g. due to a microscopic flow in the dielectricfilm. Such "faults" often re-seal due to localheating caused by the leakage current, i.e.the units are said to be "self-healing";v if the leakage current persists, the defectmay develop into a short-circuit, and the fusewill blow,v gas produced by vaporizing of themetallisation at the faulty location willgradually build up a pressure within theplastic container, and will eventually operatea pressure-sensitive device to short-circuitthe unit, thereby causing the fuse to blow.Capacitors are made of insulating materialproviding them with double insulation andavoiding the need for a ground connection.

fuse

dischargeresistor

overpressuredevice

short-circuitingcontacts

fig. E31: cross-section of a capacitorelement.

electrical characteristics

standards IEC 831, NF C 54-104, VDE 0560CSA standards, UL tests

operating rated voltage 400 Vrange rated frequency 50 Hzcapacitance tolerance 0 to + 5%temperature maximum temperature 55 °Crange average temperature over 24 h 45 °C

average annual temperature 35 °Cminimum temperature -25 °C

insulation level 50 Hz 1 mn withstand voltage: 6 kV1.2/50 µs impulse withstand voltage: 25 kV

permissible current overload standard range H range30% 50%

permissible voltage overload 10% 20%current on 400 V - 50 Hz supply 2 A/kvar 2.2 A/kvarconsumption on 230 V - 50 Hz supply 3.5 A/kvar


E

10.2 choice of protection, control devices, and connecting cables

due to the possible presence ofharmonic currents and tomanufacturing tolerances,components must be over-sized, andbased on 1.5 times rated current.

component dimensions

The choice of upstream cables and protectionand control devices depends on the currentloading.For capacitors, the current is a function of:c the applied voltage and its harmonics,c the capacitance value.The nominal current In a capacitor of kvarrating Q, supplied from a 3-phase systemhaving a phase/phase voltage of Unkilo-volts, is given by:

The permitted range of applied voltage atfundamental frequency, plus harmoniccomponents, together with manufacturingtolerances of actual capacitance (for adeclared nominal value) can result in a 50%increase above the calculated value ofcurrent.

Approximately 30% of this increase is due tothe voltage increases, while a further 15% isdue to the range of manufacturing tolerances,so that 1.3 x 1.15 = 1.5 In.All components carrying the capacitor currenttherefore, must be adequate to cover this"worst-case" condition, in an ambienttemperature of 50 °C maximum.In the case where higher temperatures (than50 °C) occur in enclosures, etc. derating ofthe components will be necessary.

In =Q

3 Un A

protection

At the instant of closing a switch to energize acapacitor, the current is limited only by theimpedances of the network upstream of thecapacitor, so that high peak values of currentwill occur for a brief period, rapidly falling tonormal operating values.This transient overcurrent however, isgenerally a high-frequency phenomenon,which is superimposed on the 50 Hz(or 60 Hz) current wave.The first peak of transient high-frequency or(sometimes) unidirectional* current has thegreatest magnitude. The maximum valueattainable, when charging an initiallyuncharged capacitor, will occur if the closingswitch contacts touch at the instant of peakpower-supply voltage.For this condition, the maximum high-frequency peak current is given by:

Where U = system phase-to-phase voltage inVoltsC = capacitance of capacitor in FaradsLo = inductance of system impedance inHenrys (system resistance is ignored).The frequency fo of the transient currentsurge is given by:

* In general, peak unidirectional currents arelower than the first peaks of high-frequencycurrents.

Ip = U2 C3 Lo

A

fo =1

2 LoCHz

Π

c for a single capacitor bank, the upstreamcables and transformers constitute thepredominant part of Lo (the systeminductance),c where a bank of capacitors is automaticallyswitched in steps, however, those units whichare already in service will initially dischargeinto an uncharged capacitor group at theinstant of switching it into service. Thetransient in-rush current from the previously-charged units will then amount to an initialpeak of

where L = the supply cable inductance inseries with each capacitorn = the number of capacitor steps alreadyenergized before closure of the switchC = capacitance of each group forming 1 step(all steps are electrically identical).The frequency f'o of the current from theenergized capacitors is given by

The total inrush current is the sum of the twoinfeeds, i.e. from the system and from thepreviously-charged bank.Generally, the frequencies of the two infeedswill not be equal.

I P UCL

' = 23

(n

n+1) A

foLC

' = 1

2ΠHz


E

10. implementation of capacitor banks (continued)

10.2 choice of protection, control devices, and connecting cables (continued)

The peak value of this transient current mustnot exceed 100 times the rated current of thecapacitors in one step of a multi-step bank(IEC 831-1). This maximum transient currentpeak occurs when the last step is energized.It is sometimes necessary to install smallseries inductors to achieve this limitation, inwhich case the manufacturer of thecapacitors should be consulted.In order to avoid undesirable nuisancetripping of controlling circuit breakers at theinstant of energizing a capacitor bank, theinstantaneous elements of overcurrenttripping relays should be given a suitably highsetting.Note: The short-circuit current-breaking rating of the circuitbreaker must be adequate to match the short-circuit levelexisting at the point of connection of the capacitor bank.

cross-sectional area

of conductors

The current rating of cables, as previouslynoted, must be based on 1.5 times thenominal current rating for the capacitor bankconcerned.

voltage transients

High-frequency voltage transientsaccompany the high-frequency currenttransients, the maximum voltage transientpeak never (in the absence of steady-stateharmonics) exceeds twice the peak value ofrated voltage, when switching an unchargedcapacitor into service.In the case of a capacitor being alreadycharged at the instant of switch closure,however, the voltage transient can attain amaximum value approaching 3 times thenormal rated peak value.This maximum condition occurs only if:c the existing voltage at the capacitor is equalto the peak value of rated voltage, andc the switch contacts close at the instant ofpeak supply voltage, andc the polarity of the power-supply voltage isopposite to that of the charged capacitor.In such a situation, the current transient willbe at its maximum possible value, viz: twicethat of its maximum when closing on to aninitially uncharged capacitor, as previouslynoted.

For any other values of voltage and polarityon the pre-charged capacitor, the transientpeaks of voltage and current will be less thanthose mentioned above. In the particular caseof peak rated voltage on the capacitor havingthe same polarity as that of the supplyvoltage, and closing the switch at the instantof supply-voltage peak, there would be novoltage or current transients.Where automatic switching of stepped banksof capacitors is considered, therefore, caremust be taken to ensure that a section ofcapacitors about to be energized is fullydischarged.The discharge delay time may be shortened,if necessary, by using discharge resistors of alower resistance value.

Section H1-2 of chapter H facilitates theselection of suitable cables, or other types ofconductor, as a function of theircharacteristics, method of installation, andambient temperature, etc.

distribution within a low-voltage installation

F

Distributionwithin a low-voltage installation

1. General1.1 the principal schemes of LV distribution F11.2 the main LV distribution board F41.3 transition from IT to TN F4

2. Essential services standby supplies2.1 continuity of electric-power supply F52.2 quality of electric-power supply F6

3. Safety and emergency-services installations,and standby power supplies3.1 safety installations F153.2 standby reserve-power supplies F153.3 choice and characteristics of reserve-power supplies F163.4 choice and characteristics of different sources F173.5 local generating sets F18

4. Earthing schemes4.1 earthing connections F194.2 definition of standardized earthing schemes F214.3 earthing schemes characteristics F234.4 choice criteria F294.5 choice of earthing method - implementation F314.6 installation and measurements of earth electrodes F32

5. Distribution boards5.1 types of distribution board F365.2 the technologies of functional distribution boards F375.3 standards F385.4 centralized control F38

6. Distributors6.1 description and choice F396.2 conduits, conductors and cables F41

7. External influences7.1 classification F477.2 protection by enclosures: IP code F49

distribution within a low-voltage installation - F1

F

* IEC 38 (1983).

1.1 the principal schemes of LV distribution

1. general

In a typical LV installation, distribution circuitsoriginate at a main general-distribution board(MGDB) from which cables are installed invarious kinds of cables-ways, conduits, etc.to supply local distribution and sub-distribution boards.The arrangement of groups of insulatedconductors and the means of fixing them andof protecting them from mechanical damage,while respecting aesthetic considerations,constitute the practical realization of anelectrical installation.

Circuit arrangementsThe creation of independent circuits todifferent parts of an installation allows:c the limitation of consequences, on thefailure of a circuit;c simplification in locating a defective circuit;c maintenance work on, or extension of acircuit may be effected without disturbing thegreater part of the installation.Division of circuits falls logically into severalcategories, each requiring an individual circuitor group of circuits, and, in some cases,particular kinds of cable (e.g. for fire-alarmand protection circuits).In general, the following circuit groups arerequired:c lighting circuits (the circuits on which themajority of insulation failures occur);c socket-outlet circuits;c heating and/or air-conditioning appliancescircuits;c power circuits for motor-driven fixed plant;c power-supply circuits for auxiliary services(indication and control);c circuits for safety systems (emergencylighting, fire-protection systems anduninterruptible-power-supplies (UPS) circuitsfor computer systems, etc.), the installation ofwhich is normally subject to strict nationalregulations and codes of practice.The most common distribution arrangementsfor low-voltage installations are described inthe following pages.

F2 - distribution within a low-voltage installation

F

1. general (continued)

1.1 the principal schemes of LV distribution (continued)

radial branched distribution schemes,in which conductor sizes areprogressively reduced at each pointof circuit sub-division are the mostcommonly used systems in mostcountries.For socket-outlet circuits in certaincountries, a ring-main circuit isstandard, in which the conductor sizeis the same throughout the circuit.Circuit wires drawn through conduits,as well as prefabricated buschannels, are commonly used.

radial branched distribution

This scheme of distribution is practicallyuniversal, and its realization generally followsarrangements similar to those illustratedbelow:

AdvantagesOne sub-divided circuit only will be isolated(by fuses or MCCB) in case of a fault.Location of the defect is simplified.

Maintenance or extensions to the circuitleaves the remainder of the installation inservice. Conductor sizes can be tapered tosuit the decreasing current levels towardsthe final sub-circuits.

DisadvantagesA fault occurring on one of the cables fromthe main distribution board will cut off supplyto all circuits of related downstreamdistribution boards and sub-distributionboards.

Conventional wiring installation (fig. F1)in buildings intended for specific use:dwellings, hotels, agricultural activities,schools, etc.

AdvantagesVirtually unrestricted passage for cableways, conduits, trays, ducts, etc.

main distributionboard

distribution board"A" worhshop

process

M M

powersub-distributionboard

lighting & heatingsub-distributionboard

fig. F1: radial branched distribution by conventional wiring at 3 levels.

With prefabricated bus channels at thesecond level of distribution (fig. F2) forindustrial and tertiary sector installations.

AdvantagesFlexibility of installation in large non-partitioned work-spaces, easy exploitation.

MGDB(main generaldistribution board)

to lightingand heatingdistributionboard

D1 D2 D3 D4

M M M

prefabricatedbus channel

a second prefabricated bus channel

process

fig. F2: radial branched distribution using prefabricated bus channels at the secondlevel of distribution.


FWith prefabricated bus-rail and pre-wiredchannels at final-circuits level (fig. F3):for offices, laboratories, etc.

AdvantagesAesthetically acceptable; flexible in locationswhere partitioning may change according toconsumers requirements; easy exploitation.


distributionboardoffice C

BA C

to heatingcontrolboard

bus railsforluminaires

prefabricatedpre-wiredcolumns,skirting-boardchannels, etc...

fig. F3: radial branched distribution using prefabricated pre-wired channels and lightingrails at final-circuits level.

simple (unbranched) radial

distribution

This scheme is used for the centralisedcontrol of an installation or process dedicatedto a particular application, its control,maintenance and surveillance.

AdvantagesA fault (other than at the busbar level) willclear one circuit only.

DisadvantagesSurplus of copper due to a multiplicity ofcircuits. Protective-device characteristicsmust be at a high level (proximity of source).


M M MM

fig. F4: simple radial distribution.


F


1.2 the main LV distribution board

The starting point for the design of anelectrical installation, and the physicallocation of distribution and sub-distributionboards, is the geographical division of theloads, shown on plans of the building(s)concerned.The HV/LV substation, standby-supply plant,and the main LV distribution board, should,for both technical and economic reasons, beplaced as near to the electrical centre of theload area as possible.

Many other factors must be consideredhowever, and in particular, the agreement ofthe power-supply authority concerning theHV/LV substation, and its related civilengineering works. In fact, very often only themain LV distribution board can be located atthe load centre, the HV/LV substation beingon the building line with the public way.

fig. F5: low-voltage main distribution board.

1.3 transition from IT to TN

In large LV installations, two voltage levelsare normally used:c one level which is generally of 380 V, 400 Vor 415 V (or exceptionally 480 V) for powercircuits, which are mainly motors;c and a second level of 220 V, 230 V or240 V (or exceptionally 277 V) for lighting andsocket-outlet circuits.The first set of voltages are the phase tophase voltages of 3-phase systems, andcorrespond respectively to the phase-to-neutral voltages given in the second set. Veryoften, particularly in factories and in somehospitals, the system comprises the threephase wires only, and operates as an ITscheme (discussed fully in Chapter GClause 6).In order to provide the lower voltages forlighting circuits, etc. while at the same time

retaining, for those circuits requiring them, theadvantages of the IT scheme, LV/LVdelta/star transformers are used, as shown infig. F6.In this way :c a 3-phase 3-wire supply is available at thesecondary side of the LV/LV transformer withphase-to-phase voltages of 220 V, 230 V or240 V as required;c all loads are connected phase-to-phaseonly (see Note);c earth faults occurring on the TN system willbe cleared rapidly by the TN system circuitbreaker and the advantages of the IT schemewill be preserved.

IT powernetwork

TN lighting, etcnetwork

residual earth fault device

PEprotectiveearthingconductor

PEprotectiveearthingconductor

fig. F6: use of a LV/LV transformer to provide a 3-phase 3-wire TN system from a 3-phase3-wire IT network.

Note: in this scheme ofdelta-connected loads, itis essential that balancedloading is maintained inall three phases.


F

2. essential services standby supplies

In order to achieve the highest possible plantperformance, it is necessary that thecontinuity and quality of the electric-powersupply be assured.

2.1 continuity of electric-power supply

continuity of power supply isachieved by:c appropriate division of theinstallation and the provision ofalternative supply sources,c the provision of local emergencystandby generation,c sub-division and duplication ofimportant circuits,c the type of earthing scheme(IT for example),c discriminative protection schemes.

A high degree of power-supply continuity canbe achieved by: dividing the installation;providing more than one source, e.g. a ring-main type service connection; automatic localstandby generation for essential services; thesub-division of circuits; the choice of earthingsystem (IT, TT, TN, etc.) and the use ofselective protection devices (fuses, relays).

the division of installations and

the provision of more than one

source

Ring-main type HV supplies, and (where theinstalled load justifies the expense) two ormore HV/LV transformers, with provision forinterconnection of the LV main distributionboards, is the most common way of ensuringa high level of supply continuity from thepower network. The use of severaltransformers allows a measure of separationof loads which would otherwise cause anunacceptable disturbance to other circuits, forexample:c computer systems which are sensitive tovoltage regulation (dips and peaks) and towaveform distortion (harmonics);c circuits which create harmonics, such asdischarge lamps, electric converters ofvarious kinds (thyristor-controlled rectifiers,inverters, motor-speed controllers, etc.);c circuits which create excessive voltagechanges, such as large motors, arc furnaces,etc.These loads and others of similarcharacteristics, i. e. loads susceptible todisturbances, and loads creating them,should preferably be supplied throughdifferent HV/LV transformers. In this way, thePCC (point of common coupling) is movedfrom the LV busbars to the HV busbars,where the effects are considerably lessbetween one group of loads and the next,and in some cases are completely eliminated.

A case in particular concerns 3rd harmonicsand all multiples of the 3rd harmonic*.If delta/star HV/LV transformers are used,then triplen harmonic currents on the LV sideof one transformer do not appear in the HV-side conductors supplying it (the currentscirculate internally around the delta winding)and so cannot affect neighbouringtransformers.Moreover, any triplen harmonic voltageswhich may be present on the HV busbars(from directly-connected HV loads forexample) will not be transformed down toLV by a delta/star transformer.The separation of loads through transformersin this way is sometimes referred to as"de-coupling".

* Known as "triplen" harmonics. Triplenharmonics are of zero-phase-sequence onbalanced 3-phase systems, which accountsfor their particular behaviour in delta/startransformers.

the provision of standby

emergency power supplies

Examples of standby emergency powersupplies include: two separate HV/LVsubstations, a privately-owned power plant,diesel-generator sets, uninterruptible staticpower supply equipment (UPS).

the sub-division of circuits

Circuits are divided into groups according totheir relative importance. In general, twogroups, commonly referred to as "essential"and "non-essential" loads are separated andsupplied from different busbars.Figure F7 shows a typical arrangement of anautomatic changeover scheme to provide LVstandby power to an "essential" loadsdistribution board.

essential loads

G

standby generator ant automatic changeovercontactor NORMAL-STANDBY

non-essential loads

HV

LV

inverter

sensitive load(computer, etc.)

fig. F7: essential and non-essential loads are separated, with automatic standby suppliesprovided for essential loads.


F

2. essential services standby supplies (continued)

A sub-group of the essential loads, namelycomputer and information technologyequipment (ITE), requires the highest degreeof continuity, stable voltage level, and qualityof wave form. These requirements are met bya static UPS inverter system.

2.1 continuity of electric-power supply (continued)

essentialloads

HV

LV

HV from a private power plant orfrom a different HV substation

non-essentialloads

essentialloads

non-essentialloads

HV

LV

Supply froma HV substation

fig. F8: an example of HV standby powersupply.

choice of earthing system

Where considerations of supply continuity areparamount, e.g. in continuous-processmanufacturing, hospital operating theatres,etc., the IT scheme* of earthing is generallyadopted.This scheme allows normal (and safe)system operation to continue in the event ofan earth-fault (by far the most common typeof insulation failure). A shutdown to trace thefault manually or automatically (see G 6.2)and effect repairs can then be carried outlater, at any convenient time (e.g. at the endof a manufacturing process, etc.).A second earth fault (if it occurs on a differentphase or on a neutral conductor) will,however, constitute a short-circuit fault, whichwill cause overcurrent relays to trip thecircuit(s).* Chapter F Sub-clause 4-5 discusses thematter of earthing schemes in more detail.

selective discrimination

by protection relays and/or fuses

The prime objective in any scheme ofautomatic protection against insulation faults,over-loading, etc., is to trip the circuit breakeror blow the fuse(s) which control(s) thefaulted circuit only, leaving all other circuitbreakers and fuses unaffected.In radial branched installations, this meansthe nearest upstream circuit breaker orfuse(s) to the fault position. All downstreamloads then being inevitably deprived ofsupply.The short-circuit (or overload) current willgenerally pass through one or more circuitbreaker(s) or fuse(s) upstream of the circuitbreaker (or fuses) controlling the faultedcable.By "discrimination" is meant that none of theupstream protective devices through whichthe fault (or overload) current flows willoperate before the protective devicecontrolling the faulted circuit has operated.In general, discrimination is achieved byincreasing the operating time of relays astheir location in a network becomes closer tothe power source. In this way, the failure tooperate of the relay closest to the fault meansthat the next relay upstream will operate in aslightly longer time.

closed open

closed

fig. F9: the principle of selectivediscrimination.

2.2 quality of electric-power supply

public and private power-supplynetworks are subject to diversedisturbances, the level and frequencyof which must be controlled andmaintained within acceptable limits.Among the most onerous are:c voltage drops, or sudden peaksand dips,c flicker,c overvoltages,c harmonic voltages and currents,particularly odd-numbered harmonics(3rd, 5th...),c high-frequency phenomena.

Power network disturbances may be of acontinuous or transitory nature. The mostimportant of these, in terms of the design andoperation of a network, are:c excessive dips (occasional voltage drops,from 15 to 90% of Un, from half a cycle to1 s) and peaks of the supply voltage atnormal frequency;c flicker, i.e. repetitive voltage drops of lessthan 10%, e.g. due to welding machines,photocopiers, etc.;c overvoltage surges;c harmonic currents and voltages, particularlythe odd harmonics (3rd, 5th...);c high-frequency phenomena.


F

the undesirable effects of voltagedips are countered in various ways,depending on the apparatus inquestion. Some common remediesinclude:c automatic load shedding and re-connection,c the use of uninterruptible power-supply units,c high-torque motors,c use of lamps which do notextinguish during dips, and othersolutions.

voltage depressions of short

duration ("dips")

Types of voltage dipAccording to the duration of the undervoltagecondition, the origin of the dip may be due toone of the following causes:c less than 0.1 second: short-circuit faultsoccurring anywhere on local LV networks,and cleared by protective devices (relays,fuses, etc.). This kind of dip is the mostcommon in "standard" systems, i.e. asopposed to networks close to heavy industry,where large disturbances are frequent;c from 0.1 to 0.5 seconds: most of the faultsoccurring on HV systems fall into thiscategory;c above 0.5 seconds: on rural networkswhere auto-reclosing circuit breakers arecommon, several successive dips may beexperienced before the fault is cleared.Other reasons for voltage dips exceeding0.5 seconds include the starting of localelectric motors (central station fire-alarmsirens produce cyclic dips in the neighbouringdistribution network, for example); lift motorswill affect local consumers, and so on...Some consequences and solutionsAmong the numerous undesirableconsequences of voltage dips, the followingmay be cited:c depending on the severity of the dip andthe type of loads in a given installation, therecan be the risk of a heavy current surgeoccurring at the restoration of normal voltage,with the consequent tripping of main circuitbreakers on overcurrent.A possible solution is a scheme of automaticload shedding and staged re-connection of

apparatuses requiring high restartingcurrents, e.g. cold incandescent lamps andresistive heating loads;c in all computer-based applications, such as:word processing, information technology,machine-tool control, and so on, voltage dipsare unacceptable, since the loss ofinformation or destruction of a programmecan occur, with catastrophic consequences.Some degree of voltage variation can betolerated and voltage-stabilizing circuits arebuilt-in for this purpose, but the universalsolution for important installations is the useof uninterruptible power supply (UPS) units,based on trickle-charged storage cells andinverters, associated with automatically-controlled diesel-generator sets;c for an electric motor, the decelerationduring a voltage dip (torque α V2) means thatits back-e.m.f. will very likely be out-of-phasewith the restored voltage. This constitutes(more or less, depending on the degree ofphase difference) conditions of short-circuit,with a corresponding heavy current flow. Incertain cases, excessive transient torquesmay occur, with a risk of damaging shafts andcouplings, etc.A common remedy is to install high-inertiahigh peak-torque motors where the drivenload allows;c some types of discharge lamp (notablymercury-vapour lamps) used for publiclighting, extinguish below a certain voltagelevel, and require several minutes (to cool)before re-igniting. The remedy is to use othertypes of lamp or to mix non-extinguishinglamps, in sufficient number to maintain a safelevel of illumination.

overvoltages

Types of overvoltageOvervoltages are distinguished in a generalway, according to their origin:c overvoltage surges due to lightning arereferred to as being of atmospheric origin.These overvoltages mainly affect overheadtransmission and distribution lines, outdoorsubstation equipment, and switchgear andtransformers, etc., connected directly to suchexposed plant.The frequency of such surges occurring isrelated to the so-called keraunic level of theregion and to the types of networkconcerned, i.e. underground cables oroverhead lines.The keraunic level is defined as the numberof days per annum on which the sound ofthunder is heard at the location concerned.c operational overvoltages.Switching at high voltage can produce surgesof voltage similar to those of atmosphericorigin; while on LV networks, the blowing offuses to clear fault current can also producerelatively severe surges of overvoltage. At HVdistribution voltage levels, switchingovervoltages are adequately suppressed bystandard lightning arresters.c overvoltage on an LV installation due tofaults on an HV system, for example:v a direct HV/LV fault occurring between theprimary and secondary windings of atransformer, or a HV line touching a LVoverhead-line distributor, etc.,v due to the flow of earth-fault current from aHV fault or lightning stroke, which passesthrough a substation earthing system that is

common to both the HV and LV networks.Methods of protection against the dangers ofsuch overvoltages are described inChapter C Sub-clause 3.1.

Consequences and solutionsAll appliances, plant, and equipment musthave a basic overvoltage withstand ability.Electric motors are particularly susceptible towinding insulation failure in the presence ofhigh-frequency high-voltage surges, whilecomputer installations and related electronicprocessing equipments are frequentlyprovided with independent (battery based)supplies, which not only assure a high qualitylevel of harmonic-free stable voltage asalready described, but effectively isolatesensitive circuits from the kind of voltagesurges in question.In industrial installations, protection againstovervoltage is considered to be achieved if allcomponents of the installation have beensuccessfully tested for power-frequencyovervoltage withstand ability, and themeasures described below have been takento protect against high-frequency high-voltageand unidirectional surge phenomena.c tests at normal power frequency.The dielectric withstand test voltage at normalpower frequency for most LV materials is2U + 1,000 volts for 1 minute (or close to thisvalue - discussions are still underway in theIEC). In IT-earthed systems a voltage-limitingdevice between the supply-transformerneutral point and earth is obligatory forprotection against power-frequency andpossible induced-surge type overvoltages.

the damaging effects of overvoltagescan be avoided:c for overvoltages at power-systemfrequency by:v assuring adequate overvoltagewithstand capability for theequipment concerned,v the use of voltage limiting deviceswhere required, in a properlyco-ordinated insulation scheme.These devices are always necessaryin IT earthed systems,c for transitory (generally impulse-type) overvoltages, by:v the application of lightningarresters,v correct coordination in theinsulation scheme noted above.


F


2.2 quality of electric-power supply (continued)

c measures against transient impulse-typeovervoltage surges.These measures depend, in addition to abasic impulse-voltage withstand capability ofthe insulating materials, on the application oflightning arresters at the origin of theinstallation, together with voltage-surgesuppression devices at sensitive points in theinstallation (e.g. at terminals of large motors).Such schemes require careful study and arebest carried out in cooperation with therelevant manufacturers.For LV installations, the transference of surgevoltages through the interwindingcapacitances of the HV/LV transformerreduces considerably the severity of theovervoltage on the LV side, compared to thaton the HV side.Transformers with earthed screens betweenHV and LV windings may also be used toprovide a costly but effective method ofeliminating the problem.v impulse voltage withstand capability ofinsulating materials.

The basic test applies a standardizedlightning voltage impulse of the form shown infigure F11, characterized by the values1.2/50 µs. These two values (in microseconds) indicate the time interval for thewave to attain its peak value from the(defined) instant of impulse initiation (i.e.1.2 µs) and the time for the impulse to fall to50% of its peak value (50 µs). These valuesare called "front time" and "time to half value"respectively.The peak value is designated by Uimp(imp = impulse).Other impulse test wave-forms, notably forrepresenting switching overvoltages, areused for test purposes, but these tests arerelevant only for very high system voltages,beyond the ranges used for distribution.Table F10 shows maximum values of peakovervoltage assumed to be possible atdifferent points in a typical LV installation.Note: Materials tested to IEC standards havean impulse withstand capability of 123% ofthe values shown in Table F10, for altitudes of0-2,000 metres.

nominal voltage of the installation levels of distribution

main distribution local final circuitsboard distribution level

board230/400 V 6 kV 4 kV 2.5 kV400/690 V 8 kV 6 kV 4 kV

table F10: assumed levels of transient overvoltage possible at different points of a typicalinstallation.

v industrial switchgear.The levels indicated in Table F12 areabstracted from IEC Publication 947. Duringthe several impulse-voltage tests, no breakdown of insulation must occur betweenphases between open contacts or betweenany phase and earth.Table F12 also includes a test for switchgear,the front face of which is insulated to class IIlevel, but at the same time including anaccessible manual-operating handle. Thisfeature provides additional safety foroperating personnel.

Note: all Compact* and Masterpact* circuitbreakers have the class II front face feature.* Merlin Gerin brand names.

kV

9,8kV

50%

1,2 50 µs

fig. F11: standardized impulse voltagewave-form 1.2/50 µs.

concerning overvoltages, IECPublication 947 takes account ofthe rules governing insulationcoordination and requires thatLV switchgear be impulse testedaccording to the withstand valuesshown in the relevant tables.

application of impulse-voltage valuesimpulse voltage circuit breakers circuit breakers/ circuit breakers/isolators

isolators + class II front facebetween phases 9.8 kV 9.8 kV 9.8 kVacross the open 9.8 kV 12.3 kV 12.3 kVcircuit breakerbetween phases and earth 9.8 kV 9.8 kV 14.7 kV

table F12: typical levels of impulse withstand voltage of industrial circuit breakers labelledUimp = 8 kV.

v use of lightning arresters.Lightning arresters are necessary (obligatoryin some countries) where an installation issupplied by a low-voltage overhead line andthe keraunic level is 25 or more. Their use is

strongly recommended, regardless of thekeraunic level, where equipment known to besusceptible to damage from overvoltagesurges is installed.


FArresters are commonly installed at each endof the LV line, generally on the first pole awayfrom the HV/LV transformer position and onthe pole at which the consumers servicecable is connected to the line. In thisarrangement the voltage will not exceed3-4.5 kV, the wave front being chopped at thislevel. The withstand value of LV componentsis normally standardized at 6 kVfor a 1.2/50 µs impulse, and equipmentscomplying with such standards are thereforeadequately protected.Assuming that the neutral conductor and thelightning arresters are connected to the sameearth electrode, then the discharge currentthrough the arresters will raise the potential ofthe neutral conductor, more or less, accordingto the resistance of the several different earthelectrodes to which it is connected.

This potential will be transferred to the phaseconductors of the installation, as described inchapter C sub-clause 3.1.If the installation earth electrode is beyondthe zone of influence of the arresterselectrode, then a TT-earthed system iscommonly used. If not, then the more-costlyTN-C-S system and equipotential "cage"earthing scheme will be necessary.

harmonic voltages and currents

Sources and types of harmonicsThe principal sources of harmonics are:c electromagnetic machines and devices,such as: iron-cored inductances, transformers(magnetizing currents), motors andgenerators and so on, and results from thenon-linear relationship between current andmagnetic flux produced by the current inferro-magnetic materials.This non-linearity produces odd-orderharmonics (principally 3rd order) with someadditional harmonics from rotating machines,which are related to winding slots in themagnetic circuits (slot ripple);c computer installations;c discharge lamps and ballasts (both lampsand ballasts are highly nonlinear);c static 3-phase converters of various kinds(inverters, speed controllers for a.c. motors,rectifiers and so on) which depend onthyristor control and current-choppingtechniques. The harmonic generation isvariable according to the function, but ingeneral the 5th and 7th harmonics areprominent, while (unlike the ferromagneticsources) the 2nd harmonic may be present;c arc furnaces create a continuous spectrumof random disturbances. If the arc is d.c.,supplied through static thyristor-controlledrectifiers, then the random perturbations areof lower average amplitude and theharmonics produced by the rectifiers arerelatively significant.

ConsequencesHarmonics give rise to (among others) thefollowing consequences:c the need to oversize certain network andinstallation components:v oversizing of conductors (refer to themanufacturers of the products concerned),v oversizing of neutral conductors (of a3-phase 4-wire system) particularly fordischarge- or fluorescent-lighting circuits; forexample, a 33% 3rd-harmonic content in thecurrent of each phase produces 100%3rd-harmonic current in the neutral conductor(since 3rd harmonic currents have zero-phase-sequence on 3-phase systems, andadd arithmetically),v oversizing of alternators (e.g. in diesel-generating sets). Refer to the manufacturersof single-phase static rectifiers and invertersfor guidance. The value of the subtransientreactance of the alternator and the type ofloads are important factors,v oversizing of transformers,

v oversizing of capacitor banks,c local overheating of magnetic circuits inmotors;c possibility of resonance between networkcapacitances and inductances (ferro-resonances) or between capacitor banks andthe system source impedance (mainlyinductive).For the latter case, the manufacturer of thecapacitor banks should be able to advise onsuitable filtering arrangements.SolutionsIn general, an installation cannot tolerate asignificant percentage of harmonics: a valueof 5% maximum is a limit oftenrecommended.Reducing the harmonic content of a systemto an acceptable level consists of:c using delta/star LV/LV transformers toisolate the 3rd (and odd multiples of the3rd-harmonic);c installing filters. Filters are of two kinds:v shunt-connected, series-resonant:extremely effective for a particular harmonic(the 5th for example) and is used inassociation with others for selective filtrationof harmonic voltages, to which they present avirtual short-circuit,v damped filter: less efficient, but covering awide band of frequencies. The action of adamped filter is described in Appendix F1,v harmonic-suppression reactor, seriesconnected to a capacitor bank. The action ofa series harmonic-suppression reductor isdescribed in Appendix F1.

the undesirable effects of harmonicvoltages and currents are minimizedby:c over-sizing of components(e.g. capacitors),v increasing insulation levelsv increasing current-carryingcapabilityc isolation of a harmonic source bysupplying it through a separateHV/LV transformer,c use of harmonics filters.


F

the undesirable effects of inductive(electric or magnetic) or common-impedance coupling betweenadjacent circuits at power-systemfrequency (with its harmonics andsuperimposed high-frequencydisturbances) together with high-frequency radiated electromagneticwaves, are minimized by:c the selection of appropriatematerials,c specific studies.



electromagnetic compatibility

(EMC)

This subject concerns all cases of coupling,by common impedance and induction(electric or magnetic) at fundamentalfrequency and harmonic frequencies,together with unidirectional and H.F. surges,and radiated electromagnetic waves,provoked by normal (switching, etc.) andabnormal (system faults, lightning, etc.)operating conditions.The unifying feature of all inductivephenomena is that electric, magnetic, orradiated electromagnetic fields, or differentcombinations of any number of them, result inproducing emfs in any conducting medium intheir paths.Essential differences are as follows:c electric or magnetic fields at power-systemfrequency and its harmonics do not, for allpractical purposes, leave the spaceimmediately surrounding their point of origin,namely a charged conductor (electric field) ora current-carrying conductor (magnetic field).Moreover, the field strength in both casesvaries inversely with the distance squaredfrom the conductor*, i.e. their zone ofinfluence rapidly diminishes with distancefrom the conductor.c the amount of energy leaving a conductorin the form of an electromagnetic wavedepends on the acceleration of electrons.This is why, for example, at the instant ofswitching on a lamp, the radiation due to theinitial acceleration of electrons can be heardin a radio receiver (i.e. the switching-transientcurrent). All disturbances on power systemswhich cause electrons to accelerate, whetherin a unidirectional or oscillatory manner, willcause a radiated wave to leave the conductorand propagate through space.The higher the frequency, the greater theacceleration of electrons, and the greater theportion of energy leaving the circuit asradiation.The field strength of a propagated wavevaries inversely as the distance from theconductor, i.e. its zone of influence is muchgreater than that of the electric or magneticfields noted above.c a further difference between the abovecases is that a non-radiating electric field canbe much stronger than its associatedmagnetic field, for example, in a high-voltagehigh-impedance (low current) circuit, andvice-versa, i.e. in a low-voltage low-impedance (high current) circuit.Whereas in a radiated wave the energy in theelectric field is exactly equal to that of themagnetic field, the circuit of origin generallybeing a capacitive/inductive combinationwhere XC = XL at the frequency of naturalresonance. It may be noted that in the contextof the present discussion, the effects of aradiated wave have not, until recently, beenvery important. With the increasing use ofwalkie-talkies, mobile transmitters andcordless telephones, etc., however, thisaspect of EMC will require closer attentionthan hitherto.The emfs induced by one or more of thethree possible modes are generally of theorder of milli- or micro-volts. However, certainmodern electronic circuits have enormousamplifying power, while at the same time in

other circuits minute currents and voltagesare normal and the circuit components arecorrespondingly fragile.

equipmentIt is for the foregoing reasons that electronicequipment requires special care, and fullprotection against interference from anypropagated or direct-coupled source.Other sources, commonly causing problems,include:c "white-noise" from fluorescent and othertypes of discharge lamps;c radiation from ignition systems of internalcombustion engines;c commercial and amateur radio transmitters,radio-directed taxis, walkie-talkies, etc.;c mains-borne interference throughconductors in the installation, for example:the opening of contactor coils or circuitbreaker tripping coils.The European Directive of 3rd March 1989,concerning electromagnetic compatibility,imposes a maximum level of permittedradiation from electrical installations and theircomponent parts (the practical application ofthe methods to adopt is still being studied atthe time of publication of this guide).

earthing arrangements and

equipotential bonding-guidance

for installation contractors

The following notes have been abstractedfrom November 1993 IEC draft proposaldocuments.

Protection against electromagneticinterference (EMI)Lightning currents in a lightning protectionsystem (LPS) or in the vicinity of a buildingcan cause overvoltages in electricalinstallations of buildings by induction. This isthe case if large metal loops exist, wheredifferent electrical wiring systems for thesupply of different electrical equipment, e.g.for power supply and for informationtechnology, are installed on different routes.In practice, a very common example is theconnection of earthed conductors of powersupply cables and of cables for informationtechnology systems in a wide (extensive)mesh.Furthermore, equipotential bonding systems,building constructions or pipe systems fornon-electrical supplies, e.g. for water, gas,heating or air conditioning, can also presentsuch induction loops.When non-electrical pipe-systems or metalparts of the building construction areconnected with the equipotential bondingsystem of the building, these metal parts maycontribute to a screening effect whichreduces the induction and contributes to theprotection against electromagneticinterference.The value of the induced voltage depends onthe rate of rise (di/dt) of the lightning current,and on the size of the loop.* except in very close proximity to the conductor, where itvaries inversely as the distance cubed.


FPower cables carrying large currents with ahigh rate of rise of current (di/dt) (e.g. thestarting current of lifts or currents controlledby rectifiers) can induce overvoltages incables of information technology systems,which may influence or damage the relatedelectrical equipment.In or near rooms for medical use, electric ormagnetic fields of electrical installations mayinterfere with medical electrical equipment (anew clause for Section 710 of IEC 364,concerning such situations is currently underconsideration).Recommended measures for reduction in theeffect of induced overvoltages depend onadequate equipotential bonding, screening,physical separation, use of filters and surgesuppressors.Consideration must be given by the plannerand designer of electrical installations to thefollowing:1. Location of potential sources ofinterferences relative to sensitive equipment.2. Location of sensitive equipment relative toheavily loaded centres, busbars orequipment, e.g. lifts.3. Provision of filters and/or surgesuppressors in the circuits feeding sensitiveelectrical equipment.4. Bonding of metal enclosures andscreening.5. Adequate separation of power and signalcables and crossovers at right angles.6. Avoidance of induction loops by selectionof a common route for the wiring systems.See also item 17 of this list.7. Use of signal cables, screened and/or intwisted pairs.8. Bonding connections should be made asshort as possible.9. Wiring systems with single core conductorsshould be enclosed in bonded metalenclosures.10. Avoidance of TN-C system (see sub-clause F 4.2 and clause G5) in installationswith sensitive equipment; see figure F13.For buildings which have, or are likely tohave, significant information technologyequipment installed, consideration must begiven to the use of separate protectiveconductors (PE) and neutral conductors (N)beyond the incoming supply point, in order tominimise the possibility of over-current andEMC problems, due to the passage of neutralcurrent through signal cables (see figuresF13 and F14).11. For TN-C-S systems, there are twopossibilities depending on the arrangementfor interconnection of equipment andextraneous conductive parts within thebuilding:c avoidance of the "TN-C section" of theTN-C-S system for distribution within thebuilding, i.e. make the separation (of the PEconductor from the PEN conductor) at theorigin of the installation;c avoidance of loops between different "TN-Ssections" of the TN-C-S system within thebuilding (see figure F14).

N

L

I

I

equipment 1

equipment 2

loop

PE

PEN

L

I1I1

I1

I3

I6

I6

I6

I6

I4I5

I5

equipment 1

equipment 2

loop

fig. F13: neutral currents in aTN-S system.

fig. F14: neutral currents in aTN-C system.


F

V

I

cablefrom theantenna

telephonepower supply

MEB

waterdistrict heatinggaswaste water

U = 0I

Vcablefrom theantenna

waterdistrict heatinggaswaste water

U ≠ 0

I

I

earth electrodeembedded in the foundation

telephonepower supply

fig. F15: introduction of armoured cables and metal pipes into buildings (examples).a) a common introduction is suitable, U = 0b) introduction at different places is not suitable, U ≠ 0.



12. Cables and pipes (e.g. for water, gas orheating) for feeding the building should enterthe building at the same place.Bonding of metal sheaths, screens and metalpipes and connections of these parts with themain equipotential bonding (MEB) of thebuilding (see figure F15).13. Avoidance of potential differencesbetween different areas of equipotentialbonding should be achieved by the use ofmetal-free fibre optic cable or other non-conducting interconnecting systems such asmicrowave or laser links.Note: the problem of earth differentialvoltages on large public telecommunicationnetworks are the responsibility of the networkoperator, who may employ other methods.

Provisions for electromagneticcompatibility (EMC)Signal connectionsIn buildings which include a PEN conductor,or where there are EMC problems on signalcables due to inadequate EMC provisions inthe electrical installations, the followingmethods may be considered to avoid orminimise the problem.14. Use of fibre optic links for signalconnections.15. Use of Class II equipment.16. Use of local transformers with separatewindings (double wound transformers) for thesupply of the information technologyequipment, taking into account therequirements of IEC 364-3, sub-clause312.2.3 and IEC 364-4, sub-clause 413.1.5,for IT* systems (local IT* systems), or ofclause 413-5, for protection by electricalseparation (e.g. transformers according toIEC 742).17. Use of suitable wiring (cabling) routing inorder to minimise the enclosed area ofcommon loops formed by the supply cablesand signal cables.* not to be confused with Information Technology. ITearthing systems are defined in sub-clause F 4.2.

Information Technology equipment may besubject to malfunction due to currents andvoltages induced in equipment or betweeninterconnected equipment.Further examples of basic techniques used toachieve immunity to incomingelectromagnetic disturbances are:a) to provide inherent immunity in theInformation Technology equipment, eitherelectrically or by use of error correction,

b) to separate the Information Technologyequipment from the sources of disturbance,c) to provide equipotential bonding betweenequipment for the relevant range offrequencies,d) to provide a low impedance earthreference plane to minimize earth potentialdifferential voltages and provide shielding.There is a continuous range of earthing andequipotential bonding methods to achieveelectromagnetic compatibility. The followingmethods exemplify this range.

Method 1: radially connected protectiveconductors (see figure F16)This uses the normal protective conductorsassociated with the supply conductors. Theprotective conductor at each equipmentprovides a relatively high impedance path forelectromagnetic disturbances (other thanmains-borne transients) such that inter-unitsignal cables are subject to a large proportionof the incident noise. Equipment musttherefore have a high immunity to functionsatisfactorily.By providing a dedicated supply circuit andearthing system serving the InformationTechnology equipment, incident disturbancescan be much reduced. In some cases, thestar earthing point (e.g. the PE bar in therelevant distribution board) of the radiallyconnected protective and functional earthingconductors for the Information Technologyequipment may be earthed by a separatededicated insulated conductor connected tothe main earthing terminal.

Method 2: use of a local horizontalequipotential bonding system (mesh)(see figure F17)The normal protective conductors aresupplemented by equipotential bonding of thecomponents of the Information Technologysystem to a local mesh (bonding mat).Depending on the frequency and the meshspacing, this can provide a low impedanceearth reference plane for signalinterconnections between those systemcomponents in close proximity to the mesh.As with Method 1, additional immunity maybe provided by a separate dedicatedInformation Technology supply circuit andearthing system, including the bonding mesh,from other supply circuits and earthingsystems and extraneous conductive partssuch as building metalwork.


F

disturbance reference levelminimum recommended

electrostatic discharge IEC 801-2 level 3 (8 kV) level 4 (15 kV)field strength IEC 801-3 level 2 level 3

(3 V/m) (10 V/m)high speed repetitive transient IEC 801-4 level 2 level 4"bursts" (contact bounce)transient overvoltages IEC 60.2

at the origin 690 V 10 kVof the installation 400 V 7.5 kVother cases 690 V 7.5 kV

400 V 5 kVcurrent waves IEC 8/20 µs 80 A 200 A(lightning, switch closing) (in preparation)

table F18: compatibility levels for installation materials.

Method 2 may be extended where necessaryby the installation of bonding meshes onother floors. All such meshes areinterconnected by (numerous) verticalbonding conductors to minimize potentialdifferences in the meshes.Method 1 is most easily implemented,especially in existing buildings. The difficultyand cost of implementation increases throughMethod 2 and its possible extensions.However, these are more likely to provide anacceptable environment for unspecified futureInformation Technology equipment.

distribution board

ITE

PE

ITE

PE

ITE

PE

signal cables signal cables

main earthing terminalor earthing-bus-conductor

ITE - Information Technology Equipment

distribution board

ITE

PE

ITE

PE

ITE

PE

signal cables signal cables

main earthing terminalor earthing-bus-conductor

fig. F16: radially connected protective conductors.

fig. F17: local horizontal bonding mesh.

The foregoing information concerningMethods 1 and 2 has been abstracted from aNovember 1993 draft proposal for a newsection (548) of IEC 364 Part 5 Chapter 54.

In the case of particular difficulties, it may benecessary to consult specialists.For current projects, and in the absence ofmore precise information, it is recommendedthat materials be selected which satisfy therequirements indicated in table F18.


F



it is possible, within a low-voltageinstallation, to provide a supply ofHigh Quality (free from disturbances)for dedicated circuits specificallyintended to supply highly sensitiveequipments, such as computer-based appliances, etc.

High Quality Supplies

The objective is to supply sensitiveequipment (information-technology devices,cash registers, micro-processors, etc.) from asource which is free from the pollutiondiscussed above, at a reasonable cost.The diagram of figure F19 represents ascheme at the level of the main generaldistribution board.The High Quality supply is achieved bymeans of an inverter and its associatedbattery of storage cells and rectifier (charger),which is supplied, in normal circumstances,from one outgoing-way of the main generaldistribution board. Continuity of supply isassured by means of a diesel-generator setand automatic changeover switch, so that anuninterrupted power supply can bemaintained indefinitely (if personnel areavailable to top up the fuel tank) or for severalhours if the substation is unattended.

HV

LVDieselgenerator

UPS

fig. F19: example of the production ofHigh Quality power supply.


F

3. safety and emergency-services installations, and standbypower supplies

3.1 safety installations

the provision of safety andemergency installations is a legalobligation.

Safety and emergency-services installationsare governed by statutory regulations,concerning:c establishments receiving the public;c high-rise apartment blocks;c establishments in which people areemployed (offices, shops, factories, etc.).They must be provided with the means forensuring the safe evacuation of personnel,notably:c security and safety lighting;c alarms and warning systems;c automatic fire detection;c fire-extinguishing systems;c smoke evacuation;

c air compressors for the pressure-operatedfire-extinguishing system;c water pumps for re-filling the fire-extinguishing system.

Apart from the general rules noted above,there are certain projects for which the safetyregulations are related to a particular process(petro-chemical, cement works...) or services(tunnel lighting, airport runway lighting...).Note: power supplies for security lighting aredescribed in Chapter J, Sub-clause 4.6.

3.2 standby reserve-power supplies

standby reserve-power plant is aneconomic necessity in numerouscircumstances where loss of supplywould have far-reachingconsequences.

Among the many applications in which aninterruption of power supply cannot betolerated, the following may be cited:c information technology installations(protection of data concerning insurances,banking, professional practices,administrations...);c industrial processes (continuity of "feed"material for continuous processing, boilerfeed-water pumps in power stations, paperproduction, desalination plants...);c food-processing industry (refrigerationplants, egg-hatching...);c telecommunications;c scientific research;c surgical operating theatres;c ticketing, plane reservations, cashregisters...;c military.

It may be noted that where severalemergency-services standby sources exist,they can also be used as reserve-powersources, on condition that any one of them isavailable and capable of starting andsupplying all safety and emergency circuits,and that the failure of one of them does notaffect the normal functioning of the others.

fig. F20: examples of reserve power supplies: central storage battery (left) and diesel-generator sets (right).


F

requirementprogrammable controllers interruptible continuousIT equipment sequential processtelecommunications process

applicationsapplications - data banks - cold-working - indications andtypes - process control and sequence control of the process

monitoring parametersexamples - IT - light - nuclearof installations services machining - chemical

banking - packaging - biologicalinsurance assembly - thermaladministration, chain - heavy mechanical- management (high inertia)systems forproduction processes

conditionsallowable zero c cduration i 1 s cof break i 15 s c (1)

i 15 mn c (1)autonomy 10 mn c (2)of source 20 mn c cminimum 1 h c c cand preferred permanent if economicalsolutionstechnique inverter with or without no-break generator permanentemployed a generator to take over or start-up generation

load of the inverter and take over setload froman inverter

3. safety and emergency-services installations, and standbypower supplies (continued)

3.3. choice and characteristics of reserve-power supplies

Apart from perceptible (albeit very brief) cutsin power supply, imperceptible interruptionsof several milli-seconds are sufficient tointerfere with certain equipments. Aspreviously noted, UPS systems are essentialin these cases, and are used together withthe reserve-power source, to ensure theutmost security.

principal specifications

In order to satisfy the requirement ofeconomical exploitation, the followingfeatures are imperative:c supply interruption is not tolerated:v in information technology (IT) systems,v in continuous-process operations, exceptfor loads of high inertia which can tolerate aninterruption in the order of 1 second;c period for conserving data in informationtechnology (IT) systems: 10 minutes;c autonomy is desirable for reserve-powersupplies installations; it is a function of theeconomics related to exploitation beyond theminimum demanded for the safety (only) ofpersonnel.

Specifications particular to safetyinstallationsRegulations covering safety installationscontain a number of conditions to berespected concerning their electric-powersources:c duration time of an interruption: accordingto the case, the following choices areimposed:v no break,v a break of less than 1 second,v a break of less than 15 seconds;c autonomy demanded for the reserve-powersource: in general it corresponds to the timenecessary to complete all safety operationsfor persons: for example, the time toevacuate an ERP (Establishments forReceiving the Public): 1 hour minimum.In large apartment blocks, the autonomy ofthe source must be 36 hours, or more.

(1) according to economic circumstances.(2) data-storage time limit.

table F21: table showing the choice of reserve-power supply types according toapplication requirements and acceptable supply-interruption times.


F

3.4. choice and characteristics of different sources

The several possible solutions arecharacterized by their availability, i.e.immediate or delayed load pick-up time, andtheir autonomy, i.e. ability to supply the loadfor a given period without attention (refillingfuel tanks for example). It is also necessary totake account of:c constraints imposed by the installation: inparticular for specialized locations, andaccording to the source(s) used;c complementary equipment;c operational constraints, e.g. according tomanufacturers operating instructions or localstatutory regulations, etc.;c routine maintenance requirements, whichcould impose less than ideal restrictionsduring the periods allotted to such work...

An overall review of the many possibilitiesand associated constraints often leads to anoptimum solution based on an inverterscheme associated with a standby diesel-generator set. A battery of storage cellsmaintains an uninterrupted supply during thestart-up and load pick-up time of the standbyset.

GM

(1) A motor-generator set permanently running and equipped with a heavy flywheel.On the loss of normal supply, the pick-up of load generally requires less than 1 second.(2) Longer if the battery is of the open type.(3) Before requiring an important overhaul.(4) A study of safety requirements allows the definition of an optimal scheme.(5) According to whether the set is pre-heated or not.

table F22: table of characteristics of different sources.

emergency and/orreserve powersupply

battery inverter cold-start diesel load generators ingenerator pick-up (1) permanent service

time required to supply loadzero time (no break) c c c1 second c1 to 10 minutes (5) ctotal time for a changeover operationzero c c crelated to the automatic c c cchangeover schemeadopted for each sourceinstallation constraints

Special location None. Unless Special location (vibrations noise nuisance, access required for(type of battery). batteries are open maintenance, fire protection).Special d.c. type. Fuel tanks.network.

additional equipment (apart from protection and changeover devices)Charger. None. Unless Starter, Inertial fly wheel AutomaticRegulator, additional batteries by batteries or and clutch. synchronizingindications and are required. compressed air. equipment.meters.

operational mode and constraintsSpecial network. Automatic. Manual or Automatic. Permanent operatingSystem losses. automatic. Fixed maximum staff.Frequent checking. Periodic startups load.

other parametersmaintenance Periodic shut-downs None. Unless open- Periodic checks, but Minor mechanical Periodic checks, but

for checking and type batteries. minimal wear and very constraints only except on minimal wear and verymaintenance work. little upkeep required. clutch and coupling shaft little upkeep required.

life expectancy (3) 4 to 5 years (2). 4 to 5 years (for 1.000 to 10.000 hrs 5 to 10 years. 10.000 hrs or 1 year.sealed batteries). and 5 to 10 years.

necessary x 2 if installation is typically 2 for 1 batteries x 2. x 2 where security x 2 if the installationredundance (4) permanent. and 3 for 2. is important. is permanent.reliability (4) constant checking Integrated checks. Mechanical and Mechanical particularly Mechanical and

is important (numerous starter clutch assembly and system ofhuman errors). batteries. coupling shaft. synchronization.


F

3. safety and emergency-services installations, and standbypower supplies (continued)

3.5 local generating sets

the association of an inverter andlocal generating set is the optimumsolution for ensuring a longautonomy.

In certain installations a power supply,independent of the normal public service, isneeded so that a local generator (usuallydriven by a diesel engine) is provided, and isassociated with an inverter.In this case the autonomy of the inverter, i.e.of the battery must be sufficient to cover theperiod of starting the diesel and coupling thegenerator to the load.The time required to effect a changeover from

one source to the other depends on thecharacteristics of the particular installation,such as: start-up sequence for the engine,possible shedding of inessential loads...The coupling is generally carried out at the LVmain general distribution board, by means ofan automatic changeover panel, an exampleof which is shown diagrammatically infig. F23.

protection and distributionequipment (complementary)

possibletransformater *

network 2

staticchangeoverswitch

batterycharger

inverterbatteryprotectionbox

network 1

diesel generator

normal powersource

protection anddistribution equipment (complementary)

manual by-passmaintenance switch

fig. F23: example of an inverter/generating-set changeover scheme, taken from MerlinGerin "Guides Pratiques".

In normal operation of the inverter, a.c. powerpasses into the rectifier section, and a verysmall part of the d.c. power at the output ofthe rectifier maintains the battery in a fully-charged condition. The remainder of the d.c.power is converted into interference-free a.c.power for the load.In the event of a changeover from normal toreserve-power generator supply, it isimportant (particularly if the load to besupplied from the generator is large, relativeto its rating) that damaging transient torqueson the generator shaft and couplings beavoided. Such torques occur for suddenly-applied loads and are due to the oscillatingtransient torque of the shaft and the steadyload torque adding and subtracting at thenatural frequency of the shaft oscillations. Inorder to avoid this phenomenon, the rectifieris controlled electronically to pass a lowcurrent initially, gradually increasing thecurrent until the load is taken entirely by thegenerator and the battery is receiving its

maintaining trickle charge. This operationlasts for 10-15 seconds.Closing down of the inverter is also carriedout progressively by similar controls on therectifier circuits.A gradual application of load also avoids thepossibility of large transient currents, andfluctuations in frequency, the latter being dueto inertia in the speed-regulation governorsystem of the prime mover.The rectifier in the conversion system createsharmonic currents which generally meansthat the reserve-power generator has to bederated (i.e. an oversized generator mayhave to be installed). This question should bediscussed with the UPS equipmentmanufacturer.In the example shown in fig. F23, the outputfrom the inverter is in synchronism with theinput supply to the rectifier, so that, in theevent of overloading or failure of the inverter,instantaneous closure of the staticchangeover switch will maintain supply.

* necessary in some cases, e.g. to adapt the voltages.


F

4. earthing schemes

4.1 earthing connections

in a building, the connection to anearth electrode and theinterconnection (bonding) of all metalparts of the building and all exposedconductive parts of electricalequipment prevents the appearanceof dangerously high voltagesbetween any two simultaneouslyaccessible metal parts.

definitions

The following terms are commonly used inindustry and in the literature. Bracketednumbers refer to fig. F24.c earth electrode (1): a conductor or groupof conductors in intimate contact with, andproviding an electrical connection with Earth(see F4.6);c earth: the conductive mass of the Earth,whose electric potential at any point isconventionally taken as zero;c electrically independent earthelectrodes: earth electrodes located at sucha distance from one another that themaximum current likely to flow through one ofthem does not significantly affect the potentialof the other(s);c earth electrode resistance: the contactresistance of an earth electrode with theEarth;c earthing conductor (2): a protectiveconductor connecting the main earthingterminal (6) of an installation to an earthelectrode (1) or to other means of earthing(e.g. TN systems);c exposed-conductive-part (see table F25):a conductive part of equipment which can betouched and which is not a live part, butwhich may become live under faultconditions,

c protective conductor (3): a conductorused for some measures of protectionagainst electric shock and intended forconnecting together any of the followingparts:v exposed-conductive-parts,v extraneous-conductive-parts,v the main earthing terminal,v earth electrode(s),v the earthed point of the source or anartificial neutral;c extraneous-conductive-part (see tableF25): a conductive part liable to introduce apotential, generally earth potential, and notforming part of the electrical installation (4).For example:v non-insulated floors or walls, metalframework of buildings,v metal conduits and pipework (not part of theelectrical installation) for water, gas, heating,compressed-air, etc. and metal materialsassociated with them;c bonding conductor (5): a protectiveconductor providing equipotential bonding;c main earthing terminal (6): the terminal orbar provided for the connection of protectiveconductors, including equipotential bondingconductors, and conductors for functionalearthing, if any, to the means of earthing.

connections

The main equipotential bonding systemThe bonding is carried out by protectiveconductors and the aim is to ensure that, inthe event of an incoming extraneousconductor (such as a gas pipe, etc.) beingraised to some potential due to a faultexternal to the building, no difference ofpotential can occur between extraneous-conductive-parts within the installation. Thebonding must be effected as close aspossible to the point(s) of entry into thebuilding, and be connected to the mainearthing terminal (6).However, connections to earth of metallicsheaths of communications cables requirethe authorization of the owners of the cables.

Supplementary equipotential connectionsThese connections are intended to connectall exposed-conductive-parts and allextraneous-conductive-parts simultaneouslyaccessible, when correct conditions forprotection have not been met, i.e. the originalbonding conductors present an unacceptablyhigh resistance.

Connection of exposed-conductive-partsto the earth electrode(s)The connection is made by protectiveconductors with the object of providing a low-resistance path for fault currents flowing toearth.

branchedprotectiveconductorsto individualconsumers (3)

extraneousconductiveparts (4)

3

3

3

mainprotectiveconductor

1

27

6

5

5

5

4

4

heating

water

gas

possible TNconnection

fig. F24: an example of a block of flats inwhich the main earthing terminal (6)provides the main equipotentialconnection. The removable link (7) allowsan earth-electrode-resistance check.


F

component parts to consider component parts to consideras exposed-conductive-parts as extraneous-conductive-parts1. cable ways 1. elements used in building constructionc conduits c metal or re-inforced concrete (RC):c impregnated-paper-insulated lead-covered v steel-framed structure,cable, armoured or unarmoured v re-inforcement rods,c mineral insulated metal-sheathed cable v prefabricated RC panels,(pyrotenax, etc.) c surface finishes:2. switchgear v floors and walls in re-inforced concretec withdrawable section without further surface treatment,3. appliances v tiled surface,c exposed metal parts of class 1 insulated c metallic covering,appliances v metallic wall covering,4. non-electrical elements 2. building services elements other thanv metallic fittings associated with cable ways electrical(cable trays, cable ladders, etc.) c metal pipes, conduits, trunking, etc. for gas,c metal objects: water and heating systems, etc.,v close to aerial conductors or to busbars c related metal components (furnaces, tanks,v in contact with electrical equipment. reservoirs, radiators),

c metallic fittings in wash rooms, bathrooms,toilets, etc.,c metallized papers.

component parts not to be consider component parts not to be consideras exposed-conductive-parts as extraneous-conductive-parts1. diverse service channels, ducts, etc. c wooden-block floors,c conduits of insulating material, c rubber-covered or linoleum-covered floors,c mouldings in wood or other insulating c dry plaster-block partition,material, c brick walls,c conductors and cables without metallic c carpets and wall-to-wall carpeting.sheaths.2. switchgearc enclosures made of insulating material.3. appliancesc all appliances having class II insulation,regardless of the type of exterior envelope.

4. earthing schemes (continued)

4.1 earthing connections (continued)

the efficient bonding and connectingto earth of all accessible metalfixtures, and all exposed-conductive-parts of electrical appliances andequipment, is essential for effectiveprotection against electric shocks.

component parts

table F25: list of exposed-conductive-parts and extraneous-conductive-parts.

installation and measurements

of earth electrodes

This subject is dealt with at the end of Sub-clause 4.6.


F

4.2 definition of standardized earthing schemes

the different earthing schemesdescribed, characterize the methodof earthing the LV neutral point of aHV/LV transformer, and the earthingof the exposed conductive-partsof the LV installation supplied from it.The choice of these methodsgoverns the measures necessary forprotection against indirect-contacthazards.

The earthing schemes to be describedcharacterize the method of earthing the LVneutral point of a HV/LV transformer (or ofany other source) and the means of earthingexposed conductive parts of the related LVinstallation.The choice of earthing scheme governs themeasures to be taken for the protection ofpersons against the hazards of indirectcontact.Several different schemes of earthing can co-exist in an installation if necessary.

TT scheme (earthed neutral)

One point* at the supply source is connecteddirectly to earth. All exposed- andextraneous-conductive-parts are connectedto a separate earth electrode at theinstallation. This electrode may or may not beelectrically independent of the sourceelectrode, the two zones of influence mayoverlap, without affecting the operation ofprotective devices.

L1L2L3N

PE

Rn

fig. F26: TT scheme.

neutral exposed-conductive-parts

Earth Earth

* Generally the star-point of a star-connected LVwinding.

TN schemes

The source is earthed as for the TT scheme(above). At the installation, all exposed- andextraneous-conductive-parts are connectedto the neutral conductor. The several versionsof TN schemes are shown below:

L1L2L3PEN

Rn

TN-C schemeThe neutral conductor is also used as aprotective conductor and is referred to as aPEN (Protective Earth and Neutral)conductor. This scheme is not permitted forconductors of less than 10 mm2 and forportable equipment.The TN-C scheme requires the establishmentof an efficient equipotential environmentwithin the installation with dispersed earthelectrodes spaced as regularly as possible.


Earth Neutral

fig. F27: TN-C scheme.

TN-S schemeThe protective conductor and the neutralconductor are separate. On underground cablesystems where lead-sheathed cables exist, theprotective conductor is generally the leadsheath. The use of separate PE and Nconductors (5 wires) is obligatory for circuits ofcross-sectional area of less than 10 mm2 forcopper and 16 mm2 for aluminium on mobileequipment.

L1L2L3NPE

Rn

fig. F28: TN-S scheme.

the TN-S (5 wires) system isobligatory for circuits of cross-sectional-area of less than 10 mm2

for copper and 16 mm2 for aluminiumon mobile equipment.

TN-C-S schemeThe TN-C and TN-S schemes can be used inthe same installation. In the scheme TN-C-Sthe TN-C (4 wires) scheme must never beused downstream of the TN-S (5 wires)scheme.

L1L2L3NPE

bad

16 mm2 6 mm2 16 mm2 16 mm2

PEN

TN-C schemenot permitted downstream of TN-S scheme

5 x 50 mm2TN-C TN-S

bad

PEPEN

fig F29: TN-C-S scheme.

The point at which the PE conductorseparates from the PEN (i.e. neutral)conductor, is generally at the origin of theinstallation.


F


4.2 definition of standardized earthing schemes (continued)

important: in the TN-C scheme theprotective conductor function of thePEN conductor takes priority. Inparticular, the PEN conductor mustbe connected directly to the earthterminal of an applicance, and thebridging connection is then made tothe neutral terminal.

L1L2L3PEN

incorrect

16 mm2 10 mm2 6 mm2 6 mm2

PEN

S < 10 mm2TNC prohibited

4 x 95 mm2TNC

correct incorrect

PEN connected to the neutralterminal is prohibited

correct

NPEN

fig. F30: connection of the PEN conductor in the TN-C scheme.


Isolated Earthor Impedance-earthed

IT scheme (isolated neutral)

No intentional connection is made betweenthe neutral point of the supply source andearth (fig. F31).Exposed- and extraneous-conductive-parts ofthe installation are connected to an earthelectrode.

R3R2R1

C3C2C1

HV/LV

Zct

HV/LV

HV/LV

Zs

fig F31: IT scheme (isolated neutral).

In practice all circuits have a leakageimpedance to earth, since no insulation isperfect. In parallel with this (distributed)resistive leakage path there is the distributedcapacitive current path, the two pathstogether constituting the normal leakageimpedance to earth (fig. F32).

Example:In a LV 3-phase 3-wire system, 1 km of cablewill have a leakage impedance due to C1,C2, C3 and R1, R2 and R3 equivalent to aneutral earth impedance Zct of 3,000 to 4,000ohms.

fig. F32: leakage impedance in anIT scheme.

fig. F33: equivalent impedance to leakageimpedances in an IT scheme.

IT scheme (impedance-earthed)

An impedance Zs (in the order of 1.000 to2.000 ohms) is connected permanentlybetween the neutral point of the transformerLV winding and earth (fig. F34).All exposed- and extraneous-conductive-parts are connected to an earth electrode.The reasons for this form of power-sourceearthing are to fix the potential of a smallnetwork with respect to earth (Zs is smallcompared to the leakage impedance) andto reduce the level of overvoltages, such astransmitted surges from the HV windings,static charges, etc. with respect to earth.It has, however, the effect of slightlyincreasing the first-fault current level(see G 3.4).

fig. F34: IT scheme (impedance-earthed).


F

4.3 earthing schemes characteristics

Each earthing scheme (often referred to asthe type of power system or system earthingarrangement) reflects three technicalchoices:c earthing method;c arrangement of PE protective conductors;c arrangement of protection against indirectcontact.These choices and their consequences willbe described for each scheme.

The consequences are related to thefollowing points:c electric shock;c fire;c power supply continuity;c overvoltages;c electromagnetic disturbances;c design and operation.

fig. F35: with a TN-S scheme, faultcurrents can be very high, limited only bythe impedance of the live conductors(phase and PEN).

TN-C schemeCharacteristicsc earthing method:v the neutral point of the transformer isconnected directly to earth and the neutralconductor is earthed at as many points aspossible,v exposed conductive parts of equipment andextraneous conductive parts are connected tothe neutral conductor;c arrangement of PE protective conductors.The PE and neutral conductors are combinedin a single PEN conductor;c arrangement of protection against indirectcontact.Given the high fault currents and touchvoltages:v automatic disconnection is mandatory inthe event of an insulation fault,v this disconnection must be provided bycircuit breakers or fuses. On installations witha combined neutral and protective conductor,residual current devices cannot be used forthis purpose since an insulation fault to earthalso constitutes a phase-neutral short-circuit.

fig. F36: any insulation fault occurringoutside a building creates a rapid rise inthe potential difference outside thebuilding. The result of a high voltageinsulation fault is shown here for a TNsystem, all earth electrodes consideredtogether.

Consequencesc earthing method:v the neutral point of the transformer isconnected directly to earth and the neutralconductor is earthed at as many points aspossible;v exposed conductive parts of equipment andextraneous conductive parts are connected tothe neutral conductor;c overvoltages:v under normal conditions, the neutral,exposed conductive parts and earth are atvirtually the same potential;v given the localised effect of earthelectrodes, the potential may vary with thedistance from the electrode. Therefore, duringa HV insulation fault, a current will flowthrough the earth electrode of the LV neutraland a power frequency voltage will appearbetween the exposed conductive parts of LVequipment and the distant earth;c power supply continuity, electromagneticcompatibility and fire: the current of insulationfaults is not limited by any earth electrodeimpedance and is therefore high (several kA).During a LV insulation fault, the supplyvoltage drop, electromagnetic disturbancesand the risk of damage (fire, motor windingsand magnetic frames) is high;c overvoltages: during a LV insulation fault,the neutral point of the triangle representingthe 3-phase voltage system is displaced andthe voltage between phase and the exposedconductive parts of the installation exceed thephase-to-neutral voltage. In practice, a valueof 1.45 Un provides a rough approximation;c protective conductors. The PE and neutralconductors are combined in a single PENconductor.

L1L2L3N

PE

Rd Rc

RmΩ

L1L2L3PEN

Rn


F


4.3 earthing schemes characteristics (continued)

The PEN conductor must satisfy therequirements of both its functions. In theevent of a conflict, the PE function haspriority.The TN-C scheme is prohibited for all circuitswith cross-sectional areas less than 10 mm2

for copper conductors or 16 mm2 foraluminium conductors. It is also prohibitedfor flexible conductors.

c fire protectionThe TN-C scheme is prohibited in premiseswhere there is a high risk of fire or explosion,for example in class BE2 and BE3 premisesrespectively for standard NFC 15-100.The reason is that the connection of theextraneous conductive parts of the building tothe PEN conductor creates a flow of currentin the structures, resulting in a risk of fire andelectromagnetic disturbances. Duringinsulation faults, these circulating currents areconsiderably increased. These phenomenaare the reason for prohibiting the use of theTN-S scheme in premises where the risk offire is high.

fig. F37: the presence of any length ofPEN conductor in a building leads to theflow of currents in the exposedconductive parts and the shielding ofequipment supplied by a TN-S scheme.

c electromagnetic compatibilityv when a PEN conductor is installed in abuilding, regardless of its length, it leads to apower frequency voltage drop under normaloperating conditions, creating potentialdifferences and therefore the flow of currentsin any circuit formed by the exposedconductive parts of the installation, theextraneous conductive parts of the building,coaxial cable and the shielding of computeror telecommunications systems.These voltage drops are amplified in moderninstallations by the proliferation of equipmentgenerating 3rd-order harmonics. Themagnitude of this harmonic is tripled in theneutral conductor instead of being cancelledout as is the case for the fundamental;

fig. F38: to determine the breaking capacity of circuit breaker C, it is necessary to knowthe impedance of the normal source, that of the replacement source and the length ofcircuit C protected by circuit breaker C.

L1L2L3NPE

Rn

v in a less apparent manner, these circulatingcurrents correspond to an imbalance of thecurrents in the distribution circuit andtherefore the creation of a magnetic field thatcan disturb cathode-ray tubes, monitors,certain medical equipment, etc. at levels aslow as 0.7 A/m (i.e. 5 A passing one metrefrom a sensitive device). This phenomenon isamplified in the event of an insulation fault;c corrosion: corrosion has two sources, firstthe DC component that the PEN conductorcan carry and second, the telluric currentsthat corrode the earth electrodes and metalstructures in the case of multiple earthing;c arrangement of protection against indirectcontact.Given the high fault currents and touchvoltages:v automatic disconnection is mandatory inthe event of an insulation fault,v this disconnection must be provided bycircuit breakers or fuses. On installations witha combined neutral and protective conductor,residual current devices cannot be used forthis purpose since an insulation fault to earthalso constitutes a phase-neutral short-circuit;c fire: protection is not provided for certaintypes of faults (impedant faults) that are notinstantly transformed into solid short-circuits.Only residual current devices offer this type ofprotection. This situation therefore presents arisk of fire;c design and operationv when using circuit breakers or fuses toprotect against indirect contact, theimpedance of the source, upstream circuitsand downstream circuits (the ones to beprotected) must be known at the designphase and subsequently remain unchangedunless the protection is also changed. Thisimpedance must be measured afterinstallation and then at regular intervals(depending on the type of premisesconcerned). The characteristics of theprotection devices are determined from theseelements;v when the installation can be supplied fromtwo sources (UPS, engine generator set,etc.), the characteristics governing theopening of the circuit breaker or the blowingof the fuse must be determined for eachconfiguration and source used;

L1L2L3NPE

incorrect incorrect

16 mm2 6 mm2 16 mm2 16 mm2

PEN

TNC schemenot permitted downstream of TN-S scheme

5 x 50 mm2TNSTNC

PENPE


F

fig. F39: with a TN-S scheme, fault currents can be very high, limited only by theimpedance of the live conductors (phase and PE).

fig. F40: with a TT scheme, fault currentsare limited by the earth electroderesistances and the accompanyingvoltage drops are very small.

L1L2L3PEN

incorrect

16 mm2 10 mm2 6 mm2 6 mm2

PEN

4 x 95 mm2TNC

incorrectcorrect correct

PEN connected to the neutralterminal is prohibited

S < 10 mm2TNC prohibited

v each circuit is designed once and for all andcannot exceed a maximum length specified indesign tables as a function of the protectiondevice used.Oversized cable cross-sections may benecessary in certain cases;v any modification to the installation requiresreassessment and checking of the protectionconditions.

TN-S scheme

Characteristicsc earthing method;v the neutral point of the transformer (or thepower supply system if the distribution uses aTN-C scheme and the installation a TN-Sscheme) is earthed just once at the upstreamend of the installation,v exposed conductive parts of equipment andextraneous conductive parts are connected tothe protection conductors which are in turnconnected to the transformer neutral;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c arrangement of protection against indirectcontact.Given the high fault currents and touchvoltages:v automatic disconnection is mandatory inthe event of an insulation fault,v this disconnection must be provided bycircuit breakers, fuses or residual currentdevices since the protection against indirectcontact can be separated from the protectionagainst phase-phase or phase-neutral short-circuits.

Consequencesc earthing method:v the neutral point of the transformer (or thepower supply system if the distribution uses aTN-C scheme and the installation a TN-Sscheme) is earthed just once at the upstreamend of the installation,v exposed conductive parts of equipment andextraneous conductive parts are connected tothe protection conductors which are in turnconnected to the transformer neutral;c overvoltages: under normal conditions, theneutral of the transformer, exposedconductive parts and earth electrode are atthe same potential, even if transientphenomena cannot be excluded and can leadto the use of lightning arrestors on thephases, neutral and exposed conductiveparts;c power supply continuity, electromagneticcompatibility and fire: the effects of HV/LVfaults, HV insulation faults and LV insulationfaults are similar to those already described

for the TN-C scheme. In particular, thecurrent of insulation faults is not limited byany earth electrode impedance and istherefore high (several kA) (see points 2,3 and 4 of the corresponding part for theTN-C scheme);c the neutral conductor cannot be earthed.This avoids creating a TN-C scheme with itsinherent disadvantages, i.e. voltage drop andload currents in the protective conductorunder normal operating conditions;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c electromagnetic compatibility:v under normal conditions, the PE conductor,as opposed to the PEN conductor, is notsubject to voltage drop and all the resultingdrawbacks of the TN-C scheme are thereforeeliminated. The TN-S scheme is similar inthis respect to the TT scheme,v in the event of an insulation fault, a highimpulse voltage appears along the PEconductor, creating the same transientproblems as for the TN-C scheme;c arrangement of protection against indirectcontact.Given the high fault currents and touchvoltages:v automatic disconnection is mandatory inthe event of an insulation fault,v this disconnection must be provided bycircuit breakers, fuses or residual currentdevices since the protection against indirectcontact can be separated from the protectionagainst phase-phase or phase-neutral short-circuits.


F



If the protection against indirect contact isprovided by overcurrent protection devices:the same characteristics apply as for theTN-C scheme;c fire: protection is not provided for impedantfaults, leading to a risk of fire;c design and operation:v calculation of the impedance of the sourcesand that of the circuit to be protected, withchecking by measurements after installationand then at regular intervals,v double determination of the disconnectionconditions when the installation can besupplied from two sources (UPS, enginegenerator set, etc.),v the circuits have a maximum length thatcannot be exceeded,v any modification to the installation requiresreassessment and checking of the protectionconditions.

TT scheme

Characteristicsc earthing method:v the neutral point of the transformer isconnected directly to earth,v exposed conductive parts of equipment areconnected by protective conductors to theearth electrode of the installation which isgenerally independent with respect to theearth electrode of the transformer neutral;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c arrangement of protection against indirectcontact.Automatic disconnection is mandatory in theevent of an insulation fault.In practice, this disconnection is carried outby residual current devices. Their operatingcurrents must be low enough for the devicesto detect the fault currents, limited by twoearth electrode resistances in series.c earthing method.

Consequencesc the neutral point of the transformer isconnected directly to earth;c exposed conductive parts of equipment areconnected by protective conductors to theearth electrode of the installation which isgenerally independent with respect to theearth electrode of the transformer neutral;c overvoltages: although, as for the TNscheme, the potential of the exposedconductive parts and the earth electrode isthe same, this may not be true for the neutralconductor which is galvanically connected toan earth electrode and the exposedconductive parts, different and in some casesrelatively far away (often the case forlightning strikes in rural areas).On industrial sites or urban areas, this is notgenerally the case. The coupling of the twoearth earth electrodes is, from an overallpoint of view, an acceptable compromise.The installation of lightning arrestors providesthe necessary level of protection;c electromagnetic compatibility: in the eventof an insulation fault, the fault current isrelatively low. For instance, with an earthelectrode resistance of 230V/100Az 2.3 Ω, the fault current is only 100 A. As aresult, the voltage drop created by the fault,

If the protection against indirect contact isprovided by residual current devices: to avoidnuisance tripping, it is often possible to usehigh residual operating currents in the orderof 1 A or more;c fire, design and operation:v the drawbacks already discussed areeliminated and we obtain the advantages ofthe TT scheme,v the use of residual current devices withoperating currents 500 mA helps to preventdamage of electrical origin which can occur inthe event of an impedant fault or due to thehigh level of insulation faults.

the accompanying electromagneticdisturbances and the transient difference inpotential between two devices (e.g. twointerconnected PCs) connected by a shieldedcable are much easier to withstand than for aTN-S scheme;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c electromagnetic compatibility: under normalconditions, the PE conductor is not subject tovoltage drop and all the resulting drawbacksof the TN-C scheme are therefore eliminated.In the event of an insulation fault, the impulsevoltage that appears along the PE conductoris low and the resulting disturbances arenegligible;c design and operation: for distributioncircuits, the cross-sectional area of the PEconductor can be less than for a TN-Sscheme;c arrangement of protection against indirectcontact:v automatic disconnection is mandatory inthe event of an insulation fault.In practice, this disconnection is carried outby residual current devices. Their operatingcurrents must be low enough for the devicesto detect the fault currents, limited by twoearth electrode resistances in series,v residual current devices are added in theform of relays for circuit breakers and in theform of RCCBs for fuses. They can protect asingle circuit or a group of circuits and theiroperating currents are chosen according tothe maximum value of the resistance R of theearth electrode for the exposed conductiveparts,v the presence of residual current devicesminimises the design and operatingconstraints. It is unnecessary to know theupstream source impedance and there is nolimit concerning the length of the circuits(except to avoid excessive voltage drop). Aninstallation can be modified or extendedwithout calculations or in-situ measurements,v the use of a replacement source by thedistribution utility or the operator is straightforward;c fire: the use of residual current devices withoperating currents i 500 mA helps preventfires of electrical origin;


F

R3R2R1

C3C2C1

HV/LV

fig. F41: with an IT scheme, fault currentsare limited by the earthing of the neutraland by the overvoltage limiter.

c electromagnetic compatibility: insulationfault currents last only a short time, less than100 ms (or less than 400 ms on distributioncircuits) and are low in magnitude.

IT scheme

Characteristicsc earthing methodThe neutral point of the transformer isisolated from earth or earthed through animpedance and an overvoltage limiter. Undernormal conditions, its potential is maintainedclose to that of the exposed conductive partsby the earth leakage capacitances of thetrunking and equipment.Exposed conductive parts of equipment andextraneous conductive parts of the buildingare connected to the building’s earthelectrode;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c arrangement of protection against indirectcontact.The fault current in the event of a singleinsulation fault is low and does not representa hazard.The occurrence of a second fault should bemade highly improbable by installing aninsulation monitoring device that will detectand indicate the occurrence of a first fault thatcan then be promptly located and eliminated.

Consequencesc earthing method.The neutral point of the transformer isisolated from earth or earthed through animpedance and an overvoltage limiter. Undernormal conditions, its potential is held closeto that of the exposed conductive parts by theearth leakage capacitances of the trunkingand equipment.Exposed conductive parts of equipment andextraneous conductive parts of the buildingare connected to the building’s earthelectrode;c overvoltages:v under normal conditions, the neutralconductor, exposed conductive parts andearth electrode are at virtually the samepotential,v an overvoltage limiter should be installed toprevent a rise in potential between the liveparts and the exposed conductive parts thatcould exceed the withstand voltage of the LVequipment in the event of a fault originating inthe high voltage installation. The protectionagainst overvoltages should be implementedaccording to the criteria common to all theearthing schemes;

c power supply continuity andelectromagnetic compatibility:v the current of a first insulation fault is low, aresult of the capacitances between the liveconductors and the expoed conductive partssuch as those of the load circuits and HFfilters,v a first low voltage insulation fault does notproduce any voltage drop on the mains orelectromagnetic disturbance over a widefrequency range corresponding to theoccurrence of a classical insulation faultcurrent.c overvoltages: after a first fault, theequipment continues to be supplied withpower and the phase-to-phase voltagegradually appears between the healthyphases and the exposed conductive parts.Equipment must be chosen with thisconstraint in mind.

Notes:c standard IEC 950 (or EN 60950) defines acategory of information processing equipmentthat can be used on IT systems;c if lightning arrestors are used, thestandards stipulate that their rated voltagesshould be chosen according to the phase-to-phase voltage.

c power supply continuity andelectromagnetic compatibility: secondinsulation fault can occur on a differentphase, creating a short-circuit and theassociated hazards. The user of an ITsystem chooses that this situation must neveroccur, even if the standards allow for thispossibility for safety reasons;c arrangement of PE protective conductors.The PE conductors are separate from theneutral conductors and are sized for thehighest fault current that can occur;c electromagnetic compatibility: under normalconditions, and even when a first insulationfault occurs, the PE conductors show novoltage drop. A high level of equipotentiality ismaintained between protective conductors,functional earthing conductors, exposedconductive parts and the extraneousconductive parts of the building to which theyare connected;c arrangement of protection against indirectcontact.The fault current in the event of a singleinsulation fault is low and does not representa hazard.


F



The occurrence of a second fault should bemade highly improbable by installing aninsulation monitoring device that will detectand indicate the occurrence of a first fault thatcan then be promptly located and eliminated.The protective devices are designed tooperate in the event of a double fault. Ifcircuit breakers or fuses are used, the rulesare similar to those for the TN scheme.Residual current devices can alos be used. Ifthe two faults occur downstream of the sameresidual current device, the device considersthe fault current as a load current and maynot trip. A separate residual current device istherefore required for each circuit.If two sites have the same installation usingan IT scheme, and their earth electrodesystems are not connected, a residual currentdevice must always be included at the headof each installation. This measure preventsone insulation fault on phase 1 of the first siteand another on phase 2 of the second sitefrom creating a dangerous situation;c fire: the use of an insulation monitoringdevice and possibly residual current deviceswith operating currents i 500 mA preventsfires of electrical origin;c design and operation:v trained maintenance personnel must beavailable for prompt locating and eliminationof the first insulation fault,v the installation must be designed with greatcare: use of the IT scheme where justified byrequirements related to continuity of supply,isolation of loads with high leakage currents(certain furnaces and certain types ofcomputer hardware), examination of theinfluence of leakage currents, in particularwith respect to residual current devices,division of the installation, etc.,v if 30 mA residual current devices are usedto protect socket circuits:- the total capacitive earth leakage currentdownstream of such a device must notexceed 10 mA. The value is estimated usingthe phase-to-phase voltage for the phase andfor the phase-to-neutral voltage for theneutral,- if the loads powered by such a circuit arenot critical, the residual current device can tripon a first insulation fault, thereby eliminating itimmediately. Otherwise the use of socketsshould be avoided or other measures taken,v comment: the earth conductor, if distributed,must be protected by 4-pole devices includingneutral protection or 2-pole devices. In finaldistribution boxes, 1-pole + neutral protectiondevices are permitted as long as the ratingsfor the phase and neutral are the same orclose, and as long as a residual currentdevice is present upstream.


F

1st criterionNo earthing scheme is universal.When it is possible to choose the earthingscheme, analyse every case separately ,basing the final choice on the specificconstraints of the electrical installation, theneeds of the user and the rules laid down byapplicable legislation or by the powerdistribution utility.The best solution often involves severaldifferent earthing schemes for differentparts of the installation.

2nd criterionThese solutions must satisfy the followingfundamental criteria:c protection against electric shock;c protection against fire of electric origin;c power supply continuity;c protection against overvoltages;c protection against electromagneticdisturbances.

3rd criterion: comparison of

earthing schemes

A comparison of the different earthingschemes leads to the followingrecommendations for use:The TT scheme is recommended forinstallations that have only limitedsurveillance or installations subject toextensions or modifications.The main reason is that it is the simplestscheme to implement in private or publicdistribution.On the other hand, given the need for twoseparate LV earthing electrodes, overvoltageprotection must often be provided.

INFLUENCE OF EARTHING ELECTRODESPrivate substation with TN schemec the inside of the equipment is not exposed(U2 = 230 V);c the incoming systems are exposed to Uf.LV consumer with TT schemec the inside of the equipment is exposed inthe event of nearby lightning strikes (Rb-If).See the section dealing with lightningarrestors;c when correctly implemented, the effects ofan HV insulation fault are eliminated.

The IT scheme is recommended if powersupply continuity is imperative.The IT scheme offers the best guaranteeconcerning the availability of power.It however requires:c a detailed study:v organisation of withstand to overvoltagesand leakage currents;c trained maintenance personnel available atall times:v to promptly eliminate any first fault,v to supervise extensions to the installation.

The TN-S scheme is recommended forinstallations that have a high level ofsurveillance or installations not subject toextensions or modifications.

This scheme is generally implementedwithout medium-sensitivity residual currentdevices.Drawbacks however include:c insulation fault currents are high and canresult in:v transient disturbances,v high risk of damage,v even fire!c a detailed study is required.If medium-sensitivity residual currentdevices are installed , they provide thisscheme with improved protection against fireand greater flexibility both in design and use.

The TN-C and TN-C-S schemes are notrecommended for use, given the risk offire and electromagnetic disturbances dueto:c voltage drops along the PEN conductors;c high insulation fault currents;c currents flowing in the extraneousconductive parts, shielding, exposedconductive parts;c uneliminated impedant faults.A detailed study is required.

The presence of any length of PEN conductorin a building leads to the flow of currents inthe exposed conductive parts and theshielding of equipment supplied by a TN-Sscheme.

4th criterionIn terms of overvoltage withstand andelectromagnetic disturbances, IT, TT andTN-S schemes are equally satisfactoryif correctly implemented .

5th criterionWhen making an economic comparison,all costs must be taken into account,including those related to:c design;c maintenance;c modification or extensions;c production losses.

4.4.1 choice criteria


F


4.4.2 comparison for each criterion

1 - level of protection against

electric shock

All earthing schemes provide equal protectionagainst electric shock as long as they areimplemented and used in accordance withapplicable standards .

2 - protection against fire of

electrical origin

For TT schemes and IT schemes , in theevent of a single fault, the insulation faultcurrent is respectively low or very low. Thesame is true for the risk of fire.For TN schemes , protection againstimpedant faults is insufficient unless residualcurrent devices are included:c in this case, it is recommended to usethe TN-S scheme together with residualcurrent devices rather than the standardTN scheme .For TN type schemes, in the event of asolid fault , the insulation fault current is highand major damage can result .The TN-C scheme presents a higher riskof fire under normal operating conditionsthan the other schemes. It is thereforeprohibited in premises presenting a high riskof fire or explosion. A load imbalance currentcirculates continuously in the PENconductors and the connected parts(e.g. metal frames, exposed conductive parts,shielding, etc.).

3 - protection against

overvoltages

For all neutral schemes, the followingsteps are necessary:1) Evaluate the disturbances to be takeninto account as a function of:c site exposure:v overvoltages due to indirect effects oflightning,v nearby direct lightning strikes;c the type of supply system:v in particular HV insulation faults;c the type of premises:v choose the appropriate level of safety.This evaluation should be carried out atpower frequency and then at higherfrequencies up to several MHz.

2) Decide on the number and quality ofequipotential zones (room, building, site) soas to organise the protection of each.In practice, for sites with a number ofbuildings supplied by the same source andinterconnected by communications media,one of the following solutions should beused :c equipotentiality by interconnection of thebuildings in one of the two following manners:v by at least one accompanying conductorwith a cross-sectional area of at least35 mm2, sized for the possible fault currents,v by a denser mesh;c complete isolation, for example by using anoptical fiber communication link without aconductive sheath.

3) Implement the necessary protection(lightning arrestors, etc.) on the lines of thedifferent incoming and outgoing electricalsystems.Comment:c the use of a TN-S scheme does noteliminate the need for the above measures;c TT installations generally require lightningarrestors (rural).Furthermore, for IT schemes, protectionagainst overvoltages due to HV faults mustbe provided by an overvoltage limiter.

4 - protection against

electromagnetic disturbances

1) For differential mode disturbances, theearthing scheme used is of noimportance .For all common mode or differential modedisturbances with frequencies greater than1 MHz, the earthing scheme used is of noimportance .

2) If correctly implemented, TT, TN-S and ITschemes can satisfy all electromagneticcompatibility criteria .Note however that for the TN-S scheme,major disturbances are produced during aninsulation fault.

3) For TN-C or TN-C-S earthing schemes, aload imbalance current circulatescontinuously in the PEN conductor,exposed conductive parts of equipmentand cable shielding . The presence of 3rdorder harmonics significantly amplifies thiscurrent in modern installations. Thiscontinuous current creates voltage dropsbetween the exposed conductive parts ofsensitive equipment connected to the PENconductor.These schemes are therefore notrecommended for use.


F

4.5 choice of earthing method - implementation

After consulting local regulations and relevantcodes of practice, etc. the tables F40 and F41can be used as an aid in deciding ondivisions and possible galvanic isolation ofappropriate sections of a proposedinstallation.

Division of sourceThis technique concerns the use of severaltransformers instead of employing one largeunit. This method has already been noted asa means of de-coupling loads, which wouldotherwise cause unacceptable disturbance toother loads, e.g. voltage depression duringthe start-up period of a large motor, and soon...The quality and continuity of supply to thewhole installation are thereby improved.

The cost of switchgear is reduced (short-circuit current level is lower). The technical/economic appraisal must be made case bycase.

Network islandsThe creation of galvanically-separated"islands" by means of LV/LV transformersallows an open choice of earthing system tobe used on the secondary side, which isindependent of any imposed earthing schemein the primary LV network. In this way, theinstallation network may be arranged foroptimum performance on different types ofload.

example

PIM

IT system

LV/LV

HV/LV

TNsystem

arc furnace

overvoltagedevice

fig. F42: a workshop in which supply continuity is paramount (IT) includes an arc furnace.The most suitable arrangement is an IT scheme for the workshop, and an isolating LV/LVtransformer to supply the arc furnace, in a TN earthing scheme.

PIMTN system

LV/LV

HV/LV

overvoltagedevice

IT system

fig. F43: a factory with a load consisting mainly of welding machines requiring a TNsystem of earthing, and a painting workshop for which supply continuity has top priority.The latter supply is shown to be provided by an IT Island system, via a LV/LV transformer.

conclusion

The optimization of the performance of thewhole installation governs the choice ofearthing system (see the followingSub-clause 4.6).Including:c initial investments, andc future operational expenditure that canarise from insufficient reliability, quality ofmaterials, safety, continuity of service, etc.which is difficult to forecast.An ideal structure would comprise:v normal power supply source,v local reserve power supply source (seeClause 3 of this chapter) and the appropriateearthing schemes.


F


4.6 installation and measurements of earth electrodes

A low-impedance earth electrode improvesconsiderably the protection of the electricalinstallation from external electromagneticinfluences, and particularly in the case ofovervoltages caused by lightning. Protectionof the building against direct lightning strokes,however, requires specialized studies, and isnot dealt with here.The quality of an earth electrode (resistanceas low as possible) depends essentially ontwo factors:c installation method;c nature of the earth.

installation methods

Three common types of installation will bediscussed:A conductor-type electrode forming a ringbeneath the perimeter of the buildingwhich houses the installation concerned(fig. F44)This solution is strongly recommended,particularly in the case of a new building.The electrode should be buried around theperimeter of the excavation made for thefoundations. It is important that the bareconductor be in intimate contact with the soil(and not placed in the gravel or aggregatehard-core, often forming a base for concrete).At least four (widely-spaced) verticallyarranged conductors from the electrodeshould be provided for the installationconnections, and where possible any re-inforcing rods in concrete work should beconnected to the electrode.The conductor forming the earth electrode,particularly when it is laid in an excavation forfoundations, must be in the earth, at least50 cm below the hard-core or aggregate basefor the concrete foundation. Neither theelectrode nor the vertical rising conductors tothe ground floor, should ever be in contactwith the foundation concrete.For existing buildings, the electrodeconductor should be buried around theoutside wall of the premises to a depth of atleast 1 metre. As a general rule, all verticalconnections from an electrode to above-ground level should be insulated for thenominal LV voltage (600-1,000 V).The conductors may be:c copper - bare cable or multiple-stripu 25 mm2;

a very effective method of obtaining alow-resistance earth connection is tobury a conductor in the form of aclosed loop in the soil at the bottomof the excavation for the buildingfoundations.The resistance R of such anelectrode (in homogeneous soil) isgiven (approximately) in ohms by:

R = 2ρ LwhereL = the length of the buried conductorin metresρ = soil resistivity in ohm-metres.

care must be taken to avoid theoccurrence of corrosion, notablywhere dissimilar metals are buried inclose proximity.

c stainless steel cable or multiple stripu 35 mm2;c galvanized-steel cable.Copper is the most expensive material, butis the most suitable from considerations ofcorrosion.The use of more than one of these materialsin the same soil is deprecated, since theelementary primary cell (e.g. zinc/copper)formed in the damp earth "electrolyte" wouldresult in problems of corrosion.In the case noted, the zinc would besacrificial to the copper, eventually leaving anuncoated (corroding) steel conductor of highsurface-to-earth contact resistance.Steel reinforcing rods in concrete, however,have approximately the same galvanicpotential, in the electro-chemical series, ascopper in soil, so that copper earth electrodesmay be connected to steel reinforcing rodswith no danger of corrosion*. Steel rods insoil, on the other hand, will corrode ifconnected to steel reinforcing rods inconcrete. Aluminium and lead are not suitablefor use as earthing electrodes.The approximate resistance R of theelectrode in ohms =2ρ LwhereL = length of conductor in metresρ = resistivity of the soil in ohm-metres (seetables F47 and F48).

Earthing rods (fig. F45)Vertically driven earthing rods are often usedfor existing buildings, and for improving (i.e.reducing the resistance of) earth electrodes incases where upper-strata soil-drying can onlybe countered by deeper penetration into theearth.The rods may be:c copper or (more commonly) copper-cladsteel. The latter are generally 1 or 2 metreslong and provided with screwed ends andsockets in order to reach considerabledepths, if necessary (for instance, the water-table level in areas of high soil resistivity)

for n rods:

R = ρ ohms nL

fig. F44: conductor buried below the levelof the foundations, i.e. not in theconcrete.* Practical experience has shown that corrosion is not aproblem at potential differences of less than 0.3 V.

c galvanized (see note below) steel pipeu 25 mm diameter, or rod u 15 mm diameter,u 2 metres long in each case. It is oftennecessary to use more than one rod, inwhich case the spacing between themshould exceed the depth to which they aredriven, by a factor of 2 to 3.The total resistance (in homogeneous soil) isthen equal to the resistance of one rod,divided by the number of rods in question.


F

for a vertical plate electrode:

R ≈ 0.8 ρ ohms L

Vertical plates (fig. F46)Rectangular plates, each side of whichu 0.5 metres, are commonly used as earthelectrodes, being buried in a vertical planesuch that the centre of the plate is at least1 metre below the surface of the soil.The plates may be:c copper of 2 mm thickness;c galvanized* steel of 3 mm thickness.The resistance R in ohms is given(approximately), by:0.8 ρ Lwhere:ρ = resistivity of the soil in ohm-metresL = the perimeter of the plate in metres* Note: Where galvanized conductingmaterials are used for earth electrodes,sacrificial cathodic protection anodes may benecessary to avoid rapid corrosion of theelectrodes where the soil is aggressive.Specially prepared magnesium anodes(in a porous sack filled with a suitable "soil")are available for direct connection to theelectrodes. In such circumstances specialistadvice is recommended.

2 mm thickness (Cu)

fig. F46: vertical plate.

data concerning earth resistivities inanalogous terrains provide a usefulbase for designing an earth-electrodesystem.

influence of the nature of the soil

nature of the terrain resistivity (in Ω.-m)swampy soil, bogs 1 - 30silt alluvium 20 - 100humus, leaf mold 10 - 150peat, turf 5 - 100soft clay 50marl and compacted clay 100 - 200jurassic marl 30 - 40sandy clay 50 - 500siliceous sand 200 - 300stoney ground 1500 - 3000grass-covered-stoney sub-soil 300 - 500chalky soil 100 - 300limestone 1000 - 5000fissured limestone 500 - 1000schist, shale 50 - 300mica schist 800granite and sandstone 1500 - 10000decomposed granite and sand stone 100 - 600

table F47: resistivity ( Ω-m) for different kinds of terrain.

nature of the terrain mean value ofresistivity (in Ω.-m)

heavy arable land, compacted humid banks 50light soil arable land, gravel, roughly banked land 500stoney soil, bare, dry sand, fissured rocks 3000

table F48: mean values of resistivity ( Ω-m) for an approximate estimation of an earth-electrode resistance with respect to zero-potential earth.

The approximate resistance R obtained inohms = ρn Lif the distance separating the rods > 4Lwhere:L = the length of the rod in metresρ = the resistivity of the soil in ohm-metres(see table F47)n = the number of rods

rods connected in parallel

L u 3 m

fig. F45: earthing rods.


F


4.6 installation and measurements of earth electrodes (continued)

measurements and constancy

of the resistance between an

earth electrode and the earth

The resistance of the electrode/earthinterface rarely remains constantAmong the principal factors affecting thisresistance are the following:c the humidity of the soil: the seasonalchanges in the moisture content of the soilcan be significant at depths of up to 2 meters.At a depth of 1 metre the value of resistivity(ρ) can vary in the ratio of 1 to 3 between awet Winter and a dry Summer in temperateregions;c frost: frozen earth can increase theresistivity of the soil by several orders ofmagnitude. This is one of the reasons,together with that noted above, forrecommending the installation of deepelectrodes;c ageing: the materials used for electrodeswill generally deteriorate to some extent forvarious reasons, for example:v chemical reactions (in acidic or alkalinesoils),

v galvanic: due to stray d.c. currents in theearth from traction systems, etc. or due todissimilar metals forming primary cells;different soils acting on sections of the sameconductor can also form cathodic and anodicareas with consequent loss of surface metalfrom the latter areas. Unfortunately, the mostfavourable conditions for low earth-electroderesistance (i.e. low soil resistivity) are alsothose in which galvanic currents can mosteasily flow.c oxidization: brazed and welded joints arethe locations at which oxidization is mostlikely to occur. Thorough cleaning of a newly-made joint, and wrapping with a suitablegreased-tape binding is the preventivemeasure commonly adopted.

Measurement of the earth-electroderesistanceThere must always be removable links whichallow the earth electrode to be isolated fromthe installation, so that periodic check tests ofthe earthing resistance can be carried out.To make such tests, two auxiliary electrodesare required, each consisting of a verticallydriven rod.c ammeter method (fig. F49)

A = RT + Rt1 = UTt1 i1

B = Rt1 + Rt2 = Ut1t2 i2

C = Rt2 + RT = Ut2T i3A + C - B = 2RTWhen the source voltage U is constant(adjusted to be the same value for each test)then:

RT = U ( 1 + 1 - 1 ) 2 i1 i3 i2In order to avoid errors due to stray earthcurrents (galvanic (d.c.) or leakage currentsfrom power and communication networks andso on) the test current should be a.c., but at adifferent frequency to that of the powersystem or any of its harmonics. Instrumentsusing hand-driven generators to make thesemeasurements usually produce an a.c.voltage at a frequency of between 85 Hz and135 Hz.The distances between the electrodes are notcritical and may be in different directions fromthe electrode being tested, according to siteconvenience. A number of tests at differ-spacings and directions are generally madefor cross-checking the test results.c use of a direct-reading earthing-resistance ohmmeter (fig. 50)These instruments use a hand-driven orelectronic-type of a.c. generator, together withtwo auxiliary electrodes, the spacing of whichmust be such that the zone of influence of theelectrode being tested should not overlap thatof the test electrode (C).

there must always be a (or a numberof) removable link(s) to isolate anearth electrode, to allow it to betested.

The test electrode (C) furthest from theelectrode (X) under test, passes a currentthrough the earth and the electrode undertest, while the second test electrode (P) picksup a voltage. This voltage, measuredbetween (X) and (P), is due to the testcurrent, and is a measure of the contactresistance (of the electrode under test) withearth. It is clear that the distance (X) to (P)must be carefully chosen to give accurateresults. If the distance (X) to (C) is increased,however, the zones of resistance ofelectrodes (X) and (C) become more remote,one from the other, and the curve of potential(voltage) becomes more nearly horizontalabout the point (O).In practical tests, therefore, the distance (X)to (C) is increased until readings taken withelectrode (P) at three different points viz: at(P) and at approximately 5 metres on eitherside of (P) give similar values. The distance(X) to (P) is generally about 0.68 of thedistance (X) to (C).

UA

t2

t1

T

fig. F49: measurement of the resistance toearth of the earth electrode of aninstallation by means of an ammeter.


F

fig. F50: measurement of the resistance to the mass of earth of electrode (X) using anearth-electrode-testing ohmmeter.

Simplified measurement (TT-system)In a TT-earthed system, a simplifiedmeasurement of the earth-electroderesistance is possible. It consists inmeasuring the impedance between the earth-electrode and the neutral conductor. It equalsthe sum of the consumer earth-electroderesistance and the distributor earth-electroderesistance. This value is always pessimistic,but the distributor earth-electrode resistanceis generally less than 5 Ω. In case of doubt,use the general method.

O

X P C

O VG

X P CV

GI

VG

voltage-drop due to the resistance of electrode (X)

voltage-drop due to the resistance of electrode (C)

a) the principle of measurement is based on assumed homogeneous soil conditionswhere the zones of influence of electrodes C and X everlap, the location of testelectrode P is difficult to determine for satisfactory results.

b) showing the effect on the potential gradient when (X) and (C) are widely spaced.The location of test electrode P is not critical and can be easily determined


F

5. distribution boards

a distribution board is among themost important elements in aninstallation. Its design andconstruction must conform with well-defined standards.

A main general distribution board is the pointat which the incoming-power supply dividesinto separate circuits, each of which iscontrolled and protected by the fuses orswitchgear of the board.In general, the power supply is connected toa set of busbars via a main switch (a circuitbreaker or switch-fuse).Individual circuits, which are usually groupedaccording to the circuit function (lighting,heating, power and so on...), are suppliedfrom the busbars. Some of the circuits feeddirectly into the busbars of local distributionboards, at which a division of circuits is made,while in extensive installations,

sub-distribution boards are sometimesnecessary, thereby creating three levels ofdistribution.Modern practice is to enclose LV distributionboards in metal housings, which afforddouble protection:c the protection of switchgear, indicatinginstruments, relays, fusegear... againstmechanical shocks, vibrations and otherexternal influences likely to interfere withoperational integrity (EMI*, dust, moisture,vermin, etc.);c the protection of personnel against thepossibility of electric shock.* electromagnetic interference

5.1 types of distribution board

Distribution boards, or an assembly of LVswitchgear, may differ according to the kind ofapplication, and to the design principleadopted (notably in the arrangement of thebusbars).

distribution boards according

to specific applications

The principal types of distribution board are:c main general distribution board(fig. F53);c local general distribution board (fig. F52);c sub-distribution board (fig. F51);c process-control i.e. "functional" distributionboard. For example MCC (motor-control-centre), heating circuits control board, and soon. The local and sub-distribution boards aredispersed throughout the installation.

The process-control boards are either:c adjacent to the main general distributionboard, orc in proximity to the process concerned.Distribution boards are generally referred toin written texts by the abbreviation DB.

the load requirements dictate thetype of distribution board to beinstalled.

fig. F51: typical sub-distribution board. fig. F52: local general distribution board.

fig. F53: an example of a large industrial main general distribution board.


F

a distinction is made between:c traditional DBs in which switch- andfusegear, etc. are fixed to a chassisat the inside-rear part of the housingc functional DBs for specificapplications.

realization of the two types of DB

Traditional DBsSwitchgear and fusegear, etc. are normallylocated on a chassis near the back of theenclosure. Indications and control devices(meters, lamps, pushbuttons, etc.) aremounted on the front face of the board.The placement of the components within theenclosure requires very careful study, takinginto account the dimensions of each item, theconnections to be made to it, and theclearances necessary to ensure safe andtrouble-free operation.A quick estimation of the area required canbe made by multiplying the sum of the areasof the individual items by 2.5.

Functional DBsDedicated to specific functions, recourse ismade to functional modules which includeswitchgear and devices, together withaccessories for mounting and connections.For example, drawer-type motor control units,which comprise contactor, fuses, isolatingswitch, control pushbuttons, indicating lamps,and so on.Design of the board is rapid, since it issufficient to add the number of modulesrequired, with vacant spaces for lateradditional units if necessary.Using these prefabricated componentsgreatly facilitates the assembly of the board.Moreover, the components of these boardshave benefited from type tests, therebyensuring an excellent safety performance.Figure F54 is an example of an industrialfunctional DB.

5.2 the technologies of functional distribution boards

There are three basic technologies in generaluse for the realization of functional DBs.

fixed functional units (fig. F54)The board is made up of fixed functional unitssuch as contactors and associated relays,according to the particular function. Theseunits are not suitable for circuit isolation (fromthe busbars for example) so that anyintervention for maintenance, modificationsand so on, requires the shutdown of theentire board.The use of removable plug-in or withdrawableunits however can minimize shutdown times,which are then limited to the interval requiredonly to remove or withdraw the unit of thecircuit concerned.

functional units which have

isolating and disconnecting

features (fig. F55)Each functional unit is mounted on aremovable panel and provided with a meansof isolation on the upstream (busbars) sideand disconnecting facilities on thedownstream (circuit) side. The complete unitcan therefore be removed for servicing,without requiring a general shutdown.

withdrawable chassis-mounted

functional units (fig. F56)The switchgear and associated accessories

fig. F56: board with withdrawable chassis-mounted units.

are mounted on a drawer-type horizontallywithdrawable chassis. The function isgenerally complex and often concerns motorcontrol.Isolation is effected on both the upstream anddownstream sides by the completewithdrawal of the unit.

fig. F54: board with fixed functional units.

fig. F55: board with isolating anddisconnecting features on each functionalunit.


F

5. distribution boards (continued)

5.3 standards

conformity with the relevant standardsis essential in order to ensure anadequate degree of operational safety.

Certains types of distribution boards (inparticular, functional distribution boards) inwhich all component parts are individuallysubject to IEC 947, also conform to specificrecommendations of IEC 439-1.

IEC standard 439-1IEC 439-1 covers LV switchgear andcontrolgear assemblies, manufactured andtype-tested as complete units.IEC 439-1 defines four "forms" of assembly,according to the degree of internalseparation, by barriers or partitions, intodifferent compartments.The separations provide:c protection against contact with live parts ofadjacent functional units;c limitation of the probability of initiatingarcing faults;c protection against the passage of foreignsolid bodies from one unit of the assembly toan adjacent unit.The following are typical forms of separationby barriers or partitions:v Form 1: no separation;v Form 2: separation of busbars from thefunctional units;v Form 3: separation of busbars from thefunctional units and separation of allfunctional units, one from another, except attheir output terminals;v Form 4: as for Form 3, but includingseparation of the outgoing terminals of allfunctional units, one from another.

The form of the separation (e.g. metallic ornon-metallic) shall be the subject ofagreement between the manufacturer andthe user.Forms 2, 3 and 4 are generally used since, ineach case, the busbars are enclosed, therebyallowing safer intervention on functional unitsor their outgoing-circuit components, thanthat afforded by Form 1.Forms 3 and 4 are adopted where the spaceavailable for each functional unit is limited, sothat without complete segregation betweenadjacent units, safe intervention formaintenance, etc., is not possible, unless thewhole distribution board is shut down.c finally, individual type tests, checks andfunctional tests carried out duringmanufacture ensure conformity to thestandard of the entire assembly.

two elements of the standardIEC 439-1 largely contribute tooperational safety:c forms of separation between adjacentfunctional units according to user'srequirementsc clearly defined individual and typetests.

fig. F57: representation of different forms of LV functional distribution boards.

5.4 centralized control

the integration of functionaldistribution boards in a system ofcentralized technical managementmust be taken into account from theearliest design stage.

The organization of data acquisition from, andinstructions to equipment, in schemes ofremote control, is assuming greaterimportance as Centralized TechnicalManagement techniques become moregeneral.In the interests of economy (incommunication-cable costs) all data andcontrol-command signals should beprocessed at the equipment (e.g. functionalDBs) in question, for transmission to andreception from the central command post.

Such signal conversions (e.g. analogue-to-digital; electrical-to-optical, etc.) to suit thedata-transmission links must therefore behoused, and supplied with suitable pollution-free power, at, or very near to the distributionboards or other equipment, concerned.

form 1 form 2 form 3 form 4


F

6. distributors

6.1 description and choice

two types of distribution are possible:c by insulated wires and cables,c by prefabricated pre-wired cablechannels.The latter are distinguished by theirease of installation, flexibility, and bythe number of connecting pointspossible.

types

Two types of distribution are possible:Distribution by insulated conductorsand cablesIncludes the mechanical protection and fixingof conduits, etc. The method of installationwill affect the maximum current permitted,as noted in IEC 439 Parts 1 and 2.

Distribution by prefabricated cablechannelsThese channels are distinguished by theirease of installation, flexibility, and by thenumber of connecting points possible.

examples

heating, etc.

general utilitiesdistribution board

main general distributionboard (MGDB)

sub-distributionboard

local generaldistributionboard

fig. F58: example 1: radial distribution wiring scheme for a hotel, using conductorsin conduits and cables.


F

6. distributors (continued)

6.1 description and choice (continued)

offices

local generaldistribution board

prefabricatedpower and light-currentdistribution column

transformer bus-ducttransformerto MGDB

MGDB

mainbusbartrunking

prefabricated pre-wiredcable channels

prefabricated distributionbusbar trunking

prefabricated pre-wiredcable channels

fig. F59: example 2: radial distribution with prefabricated bus trunking and cable channels for an entrepot installation.

selection of method-criteria

The main considerations governing thechoice of one method or the other are the firstcost and the likelihood of extensive andfrequent modifications.In the case of a fixed installation which isunlikely to be modified, either frequently orextensively, then the insulated wires-and-conduit system is the more-economicsolution.Where flexibility and ease of circuitmodification are important, then theprefabricated cable channelling systemshould be the first choice.

Design information concerning the smallestallowable (i.e. the most economic) c.s.a. ofwiring conductors and cables, for the case ofa wires-and-conduit installation, is given inSub-clauses 2.1, 2.2 and 2.3 of Chapter H1.


F

6.2 conduits, conductors and cables

IEC 364-5-52 provides information on theselection and erection of wiring systems,based on the principles described inIEC 364-1, concerning cables andconductors, their termination and/or jointing,their associated supports or suspensions,and their enclosures or methods of protectionagainst external influences.

selection of wiring systems and

methods of installation,

according to IEC 364-5-52 (1993)

The selection of a wiring system may bemade from the following table.

conductors method of installationand cables without clipped conduit cable cable cable on sup-

fixings direct trunking ducting ladder, in- port(including cable sulators wireskirting tray,trunking, cableflush floor bracketstrunking)

bareconductors - - - - - - + -insulatedconductors - - + + + - + -sheathedcables(includingarmouredandmineralinsulated)c multi-core + + + + + + 0 +c single-core 0 + + + + + 0 +

+: Permitted. -: Not permitted. 0: Not applicable, or not normally used in practice.

table F60: selection of wiring systems.

Recommended erection methods areindicated in the table below.

situations method of installationwithout with conduit cable cable cable on sup-fixings fixings trunking ducting ladder, in- port

(including cable sulators wireskirting tray,trunking, cableflush floor bracketstrunking)

building voids 21, 25 0 22, 73, - 23 12, 13, - -73, 74 74 14, 15,

16cable channel 43 43 41, 42 31, 32 4, 23 12, 13, - -

14, 15,16

buried 62, 63 0 61 - 61 0 - -in groundembedded 52, 53 51 1, 2, 5 33 24 0 - -in structuresurface - 11 3 31, 32 4 12, 13, 18 -mounted 71, 72 14, 15,

16overhead - - 0 34 - 12, 13, 18 17

14, 15,16

immersed 81 81 0 - 0 0 - -

The number in each box indicates the reference number in table H52 (IEC 364-5-52)*.-: Not permitted. 0: Not applicable, or not normally used in practice.Note: for current-carrying capacity see IEC 364-5-523.

table F61: erection of wiring systems.

* Table H52 of IEC 364-5-52 fills seven pages. Two of these pages are reproduced below by way of example.


F


6.2 conduits, conductors and cables (continued)

example description reference1 2 3

insulated conductors in conduits embedded 1in thermally insulating walls

multicore cables in conduits embedded 2in thermally insulating walls

insulated conductors in surface mounted conduits 3

single or multicore cables in surface mounted conduits 3A

insulated conductors in cable ducting on a wall 4

single or multicore cables in cable ducting on a wall 4A

insulated conductors in conduits embedded in masonry 5

single or multicore cables in conduits embedded 5Ain masonry

sheathed and/or armoured cables or sheathed singleor multicore armoured cables

c on a wall 11

c on a ceiling 11A

c on unperforated trays 12

c on perforated trays 13

c on brackets run horizontally or vertically 14

c on cleats, spaced from a wall or a ceiling 15

c on ladders 16

room

room


Fexample description reference1 2 3

sheathed single-core or multicore cables suspended 17from or incorporating suspension wire

bare or insulated conductors on insulators 18

table F62: some examples of installation methods.Note: the illustrations are not intended to depict actual product or installation practices but are indicative of the methoddescribed.


F

Designation of conduits according to themost recent IEC recommendations

new designation code

obligatory marking code1st numeral:mechanical propertiesaverage mechanical constraints:very light 1light 2medium 3high 4very high 52nd and 3rd numerals:classification according to temperature withstandcapabilities: conduit class:-5°C 05-25°C 25+90°C 90complementary marking code:1st complementary numeral:malleability of conduits:rigid (slight bends only are possible) 1malleable ("bendable") 2transversally flexible (will flatten when bent) 3flexible 42nd complementary numeral:electrical properties of conduits:with electrical continuity 1intended for use as complementary insulation 2intended for use as complementary insulationbut including electrical continuity 33rd complementary numeral: resistance of conduitsto the penetration of water, including:rain water 3projections of water (wind-blown rain) 4jets of water (from hose pipe, etc.) 5sea spray 6temporary immersion 7long-term immersion 84th complementary numeral: resistance to the penetration of solid bodies: conduits providing protection against:solid bodies greater than 2.5 mm 3solid bodies greater than 1 mm 4dust 5dust-proof (total exclusion) 65th complementary numeral: resistance to corrosion:conduits with protection:light, internal and external protection 1medium external, and light internal protection 2medium external and internal protection 3heavy external and light internal protection 4heavy external and medium internal protection 5heavy external and internal protection 66th complementary numeral: resistance to solar radiation:conduits with protection:low degree 1medium degree 2high degree 3reference number indicating the exterior diameter in mm16-20-25-32-40-50-63

3 2 8 6 1 2903 25

table F63: designation code for conduits according to the most recent IEC publications.




F

designation code (CENELEC)

"harmonized" cable Hcable derived from a harmonized cable Acable according to a national standard FRNservice voltage between conductors300 volts maximum 03500 volts maximum 05750 volts maximum 071,000 volts maximum 1symbols for insulation materialsethylene propylene rubber (EPR) Bnatural rubber or equivalent (Rubber) Rpolyvinylchloride (PVC) Vcross-linked polyethylene (XLPE) Xpolychloroprene (neoprene) (PCP) Nsymbol of sheath materialsethylene propylene rubber (EPR) Bnatural rubber or equivalent (Rubber) Rpolyvinylchloride (PVC) Vcross-linked polyethylene (XLPE) Xpolychloroprene (neoprene) (PCP) Nspecial constructionsflat divisible cable Hflat indivisible cable H2core metalscopper (no code)aluminium Acore symbolssolid single core (inflexible) Ucore of twisted strands (inflexible) Rflexible core, class 5 Fstandard flexible core (fixed installation) Khighly-flexible core, class 6 Hcomposition of the cablesnumber of conductors Xmultiplication sign if no green/yellow conductor is present Xsign when a green/yellow conductor is present Gcross-sectional area of conductor X

3 G 1,5R N - - FH 07

designation code for LV

conductors and cables

Definitionsc a conductor: as referred to in this sub-clause a conductor comprises a singlemetallic core in an insulating envelope.c a cable: a cable is made of a number ofconductors, electrically separated, butmechanically solid, and generally enclosed ina protective flexible sheath.c a cable-way: the term cable-way refers toconductors and/or cables together with themeans of support and protection, etc. forexample: cable trays, ladders, ducts,trenches, and so on... are all "cable-ways".DesignationMost countries have national standards ofcodification for conductors and cables. InEurope, a code has been established byCENELEC* which "harmonizes" the severalcodes of member countries, each of which isprogressively replacing its national code bythe CENELEC version. It may be noted, thatat the time of writing, certain cable types(notably XLPE insulated) have not yet beenincluded in the harmonized code. Table F64illustrates the form and significance of thedesignation code.* Comité Européen de Normalisation del'Electrotechnique.

as referred to in this sub-clausea conductor comprises a singlemetallic core in an insulatingenvelope.

a cable is made up of a number ofconductors, electrically separated,but mechanically solid, and generallyenclosed in a protective flexiblesheath.

the term cable-way refers toconductors and/or cables togetherwith the means of support andprotection, etc. for example : cabletrays, ladders, ducts, trenches, andso on... are all "cable-ways".

table F64: designation of conductors and cables according to CENELEC codefor harmonized cables.CENELEC has undertaken a project of harmonization of different national standards, with a view to facilitating exchangesbetween European countries.


F



Example of decoding: H07 RN-F 3G 1,5 :Harmonized cable - Nominal voltage450/750 V - Rubber insulated - Neoprene(PCP) sheathed - Flexible-3 conductors:1 green/yellow conductor - All conductors are1.5 mm2.

non-metallicprotectivesheathcore

core insulation

fig. F65: typical 3-core unarmoured cable.

conductors designation designation number of c.s.a.-voltageand cables according to the according to conductors mm 2 V

French national CENELEC codestandards code

cross-linked U 1000 R12N cable standards 1 to 5 1.5 - 630polyethylene (XLPE) U 1000 R2V not yetinflexible cables U 1000 RVFV harmonized 1 to 5 1.5 - 300

U 1000 RGPFV 1 to 5 1.5 - 240inflexible cables FRN 1X1X2 1 to 5 1.5 - 630with halogen-free FRN 1X1G1 1 to 5 1.5 - 630insulation (1) FRN 1X1X2Z4X2 1 to 5 1.5 - 300

FRN 1X1G1Z4G1 1 to 5 1.5 - 300flexible elastomeric- H 07 RN-F 2 to 5 1.5 - 500insulated cables FRN 07 RN-7 7 to 37 1.5 - 4PVC-insulated FRN 05VV-U 2 to 5 1.5 - 35cables FRN 05VV-R 2 to 5 1.5 - 35

H 05VV-F 2 to 5 0.75 - 2.5H 05VVH2-F 2 0.75

PVC-insulated H 07V-U 1 1.5 - 400conductors H 07V-R 1 1.5 - 400

H 07VK 1 1.5 - 240conductors with FRN 0...-U 1 1.5 - xxxhalogen-free FRN 0...-R 1 1.5 - xxxinsulation FRN 0...- 1 1.5 - xxx

table F66: commonly used conductors and cables.(1) cable of category C1 (non-fire-propagating cable).

identification marking of LV

conductors

LV wiring and cable conductors are markedeither by colouring, or by numbers. Thesemarkings, as recommended in IEC 446, aregoverned by the following three rules:c rule 1The green-and-yellow striped marking isreserved exclusively for protective conductorsPE or PEN.c rule 2Where a circuit includes a neutral conductor,it must be coloured light-blue (or marked bythe number 1 for multicore cables of morethan 5 conductors).Where a circuit does not include a neutralconductor, the light-blue conductor may beused as a phase conductor if it is included ina cable of more than one conductor.c rule 3Phase conductors may be identified by anycolour, except:v green-and-yellow,v green,v yellow,v light-blue (see rule 2).

Note: if a circuit requires a protectiveconductor, but the cable which is available forthe circuit does not include a green-and-yellow striped conductor, the protectiveconductor may be:c either a separate conductor with green-and-yellow striped insulation, orc a light-blue conductor, if the circuit does notinclude a neutral conductor, orc a black conductor, if the circuit includes aneutral conductor. In the two last cases, theconductor used must be marked by bands orgrommets of striped green-and-yellowcolours at the extremities of the cable, andalong any of its exposed lengths.

c rule 1The green-and-yellow stripedmarking is reserved exclusively forprotective conductors PE or PEN.

c rule 2Where a circuit includes a neutralconductor, it must be coloured light-blue (or marked by the number 1 formulticore cables of more than5 conductors).Where a circuit does not include aneutral conductor, the light-blueconductor may be used as a phaseconductor if it is included in a cable ofmore than one conductor.

c rule 3Phase conductors may be identifiedby any colour, except:v green-and-yellow,v green,v yellow,v light-blue (see rule 2).


F

7. external influences

Every electrical installation occupies anenvironment which presents a more-or-lesssevere degree of riskc for persons,c for the materials constituting the installation.Consequently, environmental conditionsinfluence the definition and choiceof appropriate installation materialsand the choice of protective measuresfor the safety of persons.The environmental conditions are referredto collectively as "external influences".

7.1 classification

external influences shall be takeninto account when choosing:c the appropriate measures toensure the safety of persons (inparticular in special locations orelectrical installations)c the characteristics of electricalequipment, such as IP degree,mechanical withstand, water ingress.

Many national standards concerned withexternal influences include a classificationscheme which is based on, or which closelyresembles, that of the international standardIEC 364-3.This standard (IEC 364-3) devotes manypages to detailed explanations of each classof influence, and for such detail the reader isreferred to the standard. Following the IECcodification scheme given below, however, aconcise list of external influences, extractedfrom Appendix A of the IEC document, ispresented in table F67.

CodificationEach condition of external influence isdesignated by a code comprising a group oftwo capital letters and a number as follows:The first letter relates to the general categoryof external influence.A = environmentB =utilizationC = construction of buildingsThe second letter relates to the nature of theexternal influence.The number relates to the class within eachexternal influence.For example the code AC2 signifies:A = environmentAC = environment-altitudeAC2 = environment-altitude > 2,000 mNote: the codification given in this chapter isnot intended to be used for markingequipment.


F

BA capabilityBA1 ordinaryBA2 childrenBA3 handicappedBA4 instructedBA5 skilled

BB resistance

BC contact with earthBC1 none

BC2 lowBC3 frequentBC4 continuous

BD evacuationBD1 (low density/

easy exit)BD2 (low density/

difficult exit)BD3 (high density/

easy exit)

BD4 (high density/difficult exit)

BE materialsBE1 no riskBE2 fire riskBE3 explosion riskBE4 contamination

risk

B

AA ambient (°C)AA1 -60°C +5°CAA2 -40°C +5°CAA3 -25°C +5°CAA4 -5°C +40°CAA5 +5°C +40°CAA6 +5°C +60°C

AB humidity

AC altitude (m)AC1 ≤ 2000AC2 > 2000

AD waterAD1 negligibleAD2 dropsAD3 spraysAD4 splashesAD5 jetsAD6 wavesAD7 immersionAD8 submersion

AE foreign bodiesAE1 negligible

AE2 smallAE3 very smallAE4 dust

AF corrosionAF1 negligibleAF2 atmosphericAF3 intermittentAF4 continuous

AG impactAG1 lowAG2 mediumAG3 high

AH vibrationAH1 lowAH2 mediumAH3 high

AJ othermechanicalstresses

AK floraAK1 no hazardAK2 hazard

AL faunaAL1 no hazardAL2 hazard

AM radiationAM1 negligibleAM2 stray currentsAM3 electromagneticAM4 ionizationAM5 electrostaticsAM6 induction

AN solarAN1 negligibleAN2 significant

AP seismicAP1 negligibleAP2 lowAP3 mediumAP4 high

AQ lightningAQ1 negligibleAQ2 indirect

AR wind

A

CA materialsCA1 non-combustibleCA2 combustible

CB structureCB1 negligible riskCB2 fire propagation

CB3 structuremovement

CB4 flexible

C

envi

ronn

emen

tut

iliza

tion

build

ing

7. external influences (continued)

table F67: concise list of important external influences (taken from Appendix A of IEC 364-3).

7.1 classification (continued)


F

7.2 protection by enclosures: IP code

The degree of protection provided by anenclosure is indicated in the IP code,recommended in IEC 529 (1989).Protection is afforded against the followingexternal influences:

c penetration by solid bodies;c protection of persons against access to liveparts;c protection against the ingress of dust;c protection against the ingress of liquids.

IP 2 3 C H

Code letters(International Protection)

First characteristic numeral(numerals 0 to 6, or letter X)

Second characteristic numeral(numerals 0 to 8, or letter X)

Additional letter (optional)(letters A, B, C, D)

Supplementary letter (optional)(letters H, M, S, W)

Where a characteristic numeral is not required to be specified, it shall be replaced by the letter "X"("XX" if both numerals are omitted).Additional letters and/or supplementary letters may be omitted without replacement.

fig. F68: IP Code arrangement.

Note: the IP code applies to electricalequipment for voltages up to and including72.5 kV.


F

Code letters

Element Numeralsor letters

Meaning for the protection of equipment

Meaning for theprotection of persons

Firstcharacteristic

numeral

0123456

IP

Against ingress of solid foreign objects

(non-protected)u 50 mm diameteru 12.5 mm diameteru 2.5 mm diameteru 1.0 mm diameterdust-protecteddust-tight

-

Against access tohazardous parts with(non-protected)back of hand fingertoolwirewirewire

Additional letter(optional)

ABCD

-

Against access tohazardous parts withback of hand fingertoolwire

Supplementaryletter (optional)

HMSW

Supplementary information specific to:High-voltage apparatusMotion during water testStationary during water testWeather conditions

-

-

Secondcharacteristic

numeral

012345678

Against ingress of water with harmfuleffects(non-protected)vertically drippingdripping (15° tilted)sprayingsplashingjettingpowerful jettingtemporary immersioncontinuous immersion

-


7.2 protection by enclosures: IP code (continued)

Elements of the IP Code and theirmeaningsA brief description of the IP Code elements isgiven in the following chart.

fig. F69: elements of the IP Code.


FExamples of the use of letters in theIP CodeThe following examples are to explain theuse and arrangement of letters in the IPCode.IP44 - no letters, no options.IPX5 - omitting first characteristic numeral.IP2X - omitting second characteristicnumeral.IP20C - using additional letter.IPXXC - omitting both characteristicsnumerals, using additional letter.IPX1C - omitting first characteristic numeral,using additional letter.IP3XD - omitting second characteristicnumeral, using additional letter.IP23S - using supplementary letter.IP21CM - using additional letter andsupplementary letter.IPX5/IPX7 - giving two different degrees ofprotection by an enclosure against both waterjets and temporary immersion for "versatile"application.

Examples of designations with theIP Codec IP Code not using optional letters:

IP 3 4Code letters1st characteristic numeral

2nd characteristic numeral

An enclosure with this designation (IP Code)(3) - protects persons, handling tools having adiameter of 2.5 mm and greater, againstaccess to hazardous parts- protects the equipment inside the enclosureagainst ingress of solid foreign objects havinga diameter of 2.5 mm and greater.(4) protects the equipment inside theenclosure against harmful effects due towater splashed against the enclosure fromany direction.c IP Code using optional letters:

IP 2 3 C SCode letters

1st characteristic numeral

2nd characteristic numeral

Additional letterSupplementary letter

An enclosure with this designation (IP Code)(2) - protects persons against access tohazardous parts with fingers- protects the equipment inside the enclosureagainst ingress of solid foreign objects havinga diameter of 12.5 mm and greater.(3) protects the equipment inside theenclosure against the harmful effects due towater sprayed against the enclosure.(C) protects persons handling tools having adiameter of 2.5 mm and greater and a lengthnot exceeding 100 mm against access tohazardous parts (the tool may penetrate theenclosure up to its full length).(S) is tested for protection against harmfuleffects due to the ingress of water when allthe parts of the equipment are stationary(e.g. the rotor of a rotating machine).An extensive description of the numerouspossible combinations of protectiverequirements is beyond the scope of thisguide. For additional information and fulldetails of application and testingrequirements of the IP Code, the reader isreferred to IEC Publication 529 (1989).

Access to the interior of a protectiveenclosureIn the normal operating state, access doorsand removable panels provided formaintenance purposes are closed, but themajority of enclosures are provided withorifices for ventilation. Adjustments throughorifices by tools (screw drivers, box spanners,etc.) from the outside are also common, whilelimited access to some "safe" sections of anenclosure are frequently provided in the formof hand-holes under a removable plate. Suchpenetrations could, unless interiorarrangements are carefully designed toprevent it, lead to accidental contact with liveparts. Figure F70 shows IEC test probesintended to prove the adequacy of protectionagainst such dangers, and the correspondingIP Code for each probe.


F

1 A

2 B

3 C

4, 5, 6 D

50 N ± 10%

10 N ± 10%

3 N ± 10%

1 N ± 10%

access probeaddit.letter

firstnumeral

test force

test rod 2.5 mm diameter, 100 mm long

sphere 50 mm diameter

jointed test finger

test wire 1.0 mm diameter, 100 mm long

approx. 100

handle guard(insulating material)

rigid test sphere (metal)

jointed test finger (metal)

stop face (Ø50x20)insulating material

edges free from burrs

approx. 100

rigid test wire (metal)handle(insulating material)

stop face(insulating material)

Ø12

80

edges free from burrs

approx. 100

rigid test rod (metal)handle(insulating material)

stop face(insulating material)

Ø35

Ø10 Ø1100

Ø45

Ø50Ø10

4

Ø35

Ø10 Ø2.5100

fig. F70: access probes for the tests for protection of persons against access to hazardous parts.


7.2 protection by enclosures: IP code (continued)

protection against mechanical

impact

The selection of equipment according to anadequate IP code can ensure safety only ifthe enclosure is sufficiently robust to sustainanticipated mechanical stresses, notablyimpact forces, without distortion which willadversely affect its IP classification.Specifications for such equipment shouldtherefore include the appropriate AG code(AG1, AG2 or AG3) according to the severityof possible impact stresses as listed intable F67.Tests for standardized impact severities arebeing "harmonized" November 1993internationally, and are based on four levelsof impact energy, viz:

level energy in joules1 0.2552 2.03 6.04 20.0

It is recommended that these values be usedin specifications, pending quantification of theAG code in IEC 364-3.

protection against corrosion

For similar reasons to those mentionedabove (i.e. possible reduction in the degree ofprotection required, due to externalinfluences), the possibility of weakening ofenclosures or enlarging of orifices and so on,caused by corrosion must also be given dueconsideration.The severity of the corrosive environmentmay be indicated in the equipmentspecifications by the AF code (AF1, AF2, AF3or AF4) as listed in table F67, and defined insub-clause 321 of IEC 364-3.

protection against electric shocks

G

Protection against electric shocks

1. General1.1 electric shock G11.2 direct and indirect contact G1

2. Protection against direct contact2.1 measures of protection against direct contact G22.2 additional measure of protection against direct contact G3

3. Protection against indirect contact3.1 measure of protection by automatic disconnection of the supply G43.2 automatic disconnection for a TT-earthed installation G53.3 automatic disconnection for a TN-earthed installation G63.4 automatic disconnection on a second fault in an IT-earthed system G83.5 measures of protection against direct or indirect contact without circuit G10disconnection

4. Implementation of the TT system4.1 protective measures G134.2 types of RCD G144.3 coordination of differential protective devices G15

5. Implementation of the TN system5.1 preliminary conditions G185.2 protection against indirect contact G185.3 high-sensitivity RCDs G225.4 protection in high fire-risk locations G225.5 when the fault-current-loop impedance is particularly high G23

6. Implementation of the IT system6.1 preliminary conditions G246.2 protection against indirect contact G256.3 high-sensitivity RCDs G296.4 in areas of high fire-risk G296.5 when the fault-current-loop impedance is particularly high G30

7. Residual current differential devices (RCDs)7.1 description G317.2 application of RCDs G317.3 choice of characteristics of a residual-current circuit breaker G34(RCCB - IEC 1008)

G

protection against electric shocks - G1

1. general

1.1 electric shock

when a current exceeding 30 mApasses through a part of a humanbody, the person concerned is inserious danger if the current is notinterrupted in a very short time.

the protection of persons againstelectric shock in LV installations mustbe provided in conformity withappropriate national standards andstatutory regulations, codes ofpractice, official guides and circulars,etc.Relevant IEC standards include:IEC 364, IEC 479-1, IEC 755,IEC 1008, IEC 1009and IEC 947-2 appendix B.

electric shockAn electric shock is the pathophysiologicaleffect of an electric current through thehuman body.Its passage affects essentially the circulatoryand respiratory functions and sometimesresults in serious burns. The degree ofdanger for the victim is a function of themagnitude of the current, the parts of thebody through which the current passes, andthe duration of current flow.IEC Publication 479-1 defines four zones ofcurrent-magnitude/time-duration, in each ofwhich the pathophysiological effects aredescribed (fig. G1). Any person coming intocontact with live metal risks an electric shock.

Curve C1 (of figure G1) shows that when acurrent greater than 30 mA passes through apart of a human being, the person concernedis likely to be killed, unless the current isinterrupted in a relatively short time.The point 500 ms/100 mA close to the curveC1 corresponds to a probability of heartfibrillation of the order of 0.14%.The protection of persons against electricshock in LV installations must be provided inconformity with appropriate nationalstandards and statutory regulations, codes ofpractice, official guides and circulars, etc.Relevant IEC standards include: IEC 364,IEC 479-1, IEC 755, IEC 1008, IEC 1009and IEC 947-2 appendix B.

0,1 0,2 0,5 1 2 5 10 20 50 100 200 500 1000 2000 5000 10000mA

current passing through the body Is

10

20

50

100

200

500

1000

5000

10000ms

2000

21 3 4

A B C1 C2 C3

duration of current flow t

∂ imperceptible∑ perceptible∏ reversible effects: muscular contractionπ possibility of irreversible effectsC1: no heart fibrillationC2: 5% probabilityof heart fibrillationC3: 50% probabilityof heart fibrillation

fig. G1: curve C1 (of IEC 479-1) defines the current-magnitude/time-duration limits whichmust not be exceeded.

1.2 direct and indirect contact

standards and regulations distinguishtwo kinds of dangerous contact:c direct contact,c indirect contact,and corresponding protectivemeasures.

direct contactA direct contact refers to a person cominginto contact with a conductor which is live innormal circumstances.

indirect contactAn indirect contact refers to a person cominginto contact with a conductive part which isnot normally alive, but has become aliveaccidentally (due to insulation failure or someother cause).

1 2 3 N

busbars

Is

insulationfailure

Is

insulationfailure1 2 3 PE conductor

Id

fig. G2: direct contact. fig. G3: indirect contact.

Is: touch current Id: insulation fault current

G

G2 - protection against electric shocks

2. protection against direct contact

two measures of protection againstdirect-contact hazards are oftenimposed, since, in practice, the firstmeasure may not prove to beinfallible.

Two complementary measures are commonlyemployed as protection against the dangersof direct contact:c the physical prevention of contact with liveparts by barriers, insulation, inaccessibility,etc.c additional protection in the event that adirect contact occurs, despite the abovemeasures. This protection is based onresidual-current operated high-sensitivity fast-acting relays, which are highly effective in themajority of direct contact cases.

2.1 measures of protection against direct contact

IEC and national standardsfrequently distinguish betweendegrees of protectionc complete (insulation, enclosures)c partial or particular.

measures of complete protectionProtection by the insulation of live partsThis protection consists of an insulation whichconforms to the relevant standards. Paints,lacquers and varnishes do not provide anadequate protection.

Protection by means of barriers orenclosuresThis measure is in widespread use, sincemany components and materials are installedin cabinets, pillars, control panels anddistribution-board enclosures, etc. To beconsidered as providing effective protectionagainst direct-contact hazards, theseequipments must possess a degree ofprotection equal to at least IP2X or IPXXB(see Chapter F Sub-clause 7.2).Moreover, an opening in an enclosure (door,panel, drawer, etc.) must only be removable,opened or withdrawn:c by means of a key or tool provided for thepurpose, orc after complete isolation of the live parts inthe enclosure, orc with the automatic action of an interveningmetal shutter, removable only with a key orwith tools. The metal enclosure and all metalshutters must be bonded to the protectiveearthing conductor for the installation.

partial measures of protectionProtection by means of obstacles, or byplacing out of reachThis practice concerns locations to whichqualified, or otherwise authorized personnelonly, have access.

particular measuresof protectionProtection by the use of extra-low voltageSELV (Safety Extra Low Voltage) schemesThis measure is used only in low-powercircuits, and in particular circumstances, asdescribed in Sub-clause G3.5.

fig. G4: inherent direct-contact protectionby the insulation of a 3-phase cable withouter sheath.

fig. G5: example of direct-contactprevention by means of an earthed metalenclosure.

G


2.2 additional measure of protection against direct contact

an additional measure of protectionagainst the hazards of direct contactis provided by the use of residual-current operated devices, whichoperate at 30 mA or less, and arereferred to as RCDs of highsensitivity.

All the preceding protective measures arepreventive, but experience has shown that forvarious reasons they cannot be regarded asbeing infallible. Among these reasons may becited:c lack of proper maintenance,c imprudence, carelessness,c normal (or abnormal) wear and tear ofinsulation; for example, flexure and abrasionof connecting leads,c accidental contact,c immersion in water, etc. - a situation inwhich insulation is no longer effective.In order to protect users in suchcircumstances, highly-sensitive fast-trippingdevices, based on the detection of residualcurrents to earth (which may or may not bethrough a human being or animal) are usedto disconnect the power supply automatically,and with sufficient rapidity to preventpermanent injury to, or death byelectrocution, of a normally healthy humanbeing.

fig. G6: high-sensitivity RCD.

These devices operate on the principle ofdifferential current measurement, in whichany difference between the current entering acircuit and that leaving it, must (on a systemsupplied from an earthed source) be flowingto earth, either through faulty insulation orthrough contact of an earthed object, such asa person, with a live conductor.Standard residual-current devices, referred toas RCDs, sufficiently sensitive for direct-contact protection are rated at 30 mA ofdifferential current. Other standard IECratings for high-sensitivity RCDs are 10 mAand 6 mA (generally used for individualappliance protection).This additional protection is imposed incertain countries for circuits supplying socketoutlets of ratings up to 32 A, and even higherif the location is wet and/or temporary (suchas work-sites for example).Chapter L section 3 itemizes various commonlocations in which RCDs of high sensitivityare obligatory (in some countries), but in anycase, are highly recommended as aneffective protection against both direct- andindirect-contact hazards.

IEC wiring regulations impose theuse of RCDs on circuits supplyingsocket outlets, installed in particularlocations considered to be potentiallydangerous, or used for specialpurposes. Some national wiringregulations impose their use on allcircuits supplying socket outlets.

G


3. protection against indirect contact

national regulations covering LVinstallations impose, or stronglyrecommend, the provision of devicesfor indirect-contact protection.

the measures of protection are:c automatic disconnection of supply(at the first or second fault detection,depending on the system of earthing)c particular measures according tocircumstances.

Conductive material(1) used in themanufacture of an electrical appliance, butwhich is not part of the circuit for theappliance, is separated from live parts by the"basic insulation". Failure of the basicinsulation will result in the conductive partsbecoming live.Touching a normally-dead part of an electricalappliance which has become live due to thefailure of its insulation, is referred to as anindirect contact.Various measures are adopted to protectagainst this hazard, and include:c automatic disconnection of power supply tothe appliance concerned,

c special arrangements such as:v the use of class II insulation materials, or anequivalent degree of insulation,v non-conducting location(2) - out-of-reach orinterposition of barriers,v equipotential locality,v electrical separation by means of isolatingtransformers.(1) Conductive material (usually metal) which may betouched, without dismantling the appliance, is referred to as"exposed conductive parts".(2) The definition of resistances of the walls, floor and ceilingof a non-conducting location is given in Sub-clause G3.5

3.1 measure of protection by automatic disconnection of the supply

protection against indirect-contacthazards by automatic disconnectionof the supply can be achieved if theexposed conductive parts ofappliances are properly earthed.

principleThis protective measure depends on twofundamental requirements:c the earthing of all exposed conductive partsof equipment in the installation and theconstitution of an equipotential bondingnetwork (see Sub-clause F4-1),c automatic disconnection of the section ofthe installation concerned, in such a way thatthe touch-voltage/time safety requirementsare respected for any level of touch voltageUc(3).(3) touch voltage Uc:Touch voltage Uc is the voltage existing (as the result ofinsulation failure) between an exposed conductive part andany conductive element within reach which is at a different(generally earth) potential.The greater the value of Uc, the greater the rapidity ofsupply disconnection required to provide protection (seeTables G8 and G9). The highest value of Uc that can betolerated indefinitely without danger to human beings iscalled the "conventional touch-voltage limit" (UL).

installation earth electrode Uc

fig. G7: in this illustration the dangeroustouch voltage Uc is from hand to hand.

reminder of the theoretical

disconnecting-time limits*in practice the disconnecting timesand the choice of protection schemesto use depend on the kind of earthingsystem concerned; TT, TN or IT.Precise indications are given in thecorresponding paragraphs.

assumed maximum disconnectingtouch time for the protectivevoltage device (seconds)(V) alternating direct

current current< 50 5 550 5 575 0.60 590 0.45 5120 0.34 5150 0.27 1220 0.17 0.40280 0.12 0.30350 0.08 0.20500 0.04 0.10

assumed maximum disconnectingtouch time for the protectivevoltage device (seconds)(V) alternating direct

current current25 5 550 0.48 575 0.30 290 0.25 0.80110 0.18 0.50150 0.12 0.25230 0.05 0.06280 0.02 0.02

table G8: maximum safe duration of theassumed values of touch voltage inconditions where U

L = 50 V(1).

(1) The resistance of the floor and the wearing of shoes aretaken into account in these values.

* For most locations, the maximum permitted touch voltage(UL) is 50 V. For special locations, the limit is reduced to25 V. See G4-1 and Clause L3.

table G9: maximum safe duration of theassumed values of touch voltage inconditions where U L = 25 V.

G


3.2 automatic disconnection for a TT-earthed installation

automatic disconnection for aTT-earthed installation is effected bya RCD having a sensitivity of

I∆n i UL = 50 V*

RA RA

where RA = resistance of theinstallation earth electrode* 25 V in some particular cases.

principleIn this scheme all the exposed andextraneous conductive parts of the installationmust be connected to a common earthelectrode. The supply system neutral isnormally earthed at a point outside the areaof influence of the electrode for theinstallation, but need not be so.The impedance of the earth-fault looptherefore consists mainly of the two earthelectrodes (i.e. the source and installationelectrodes) in series, so that the magnitude ofthe earth-fault current is generally too small tooperate overcurrent relays or fuses, and theuse of a differential-current form of protectionis essential.This principle of protection is also valid if onecommon earth electrode only is used, notablyin the case of a consumer-type substationwithin the installation area, where spacelimitations may impose the adoption of a TNearthing scheme, but where all otherconditions required by the TN system cannotbe fulfilled.

Automatic protection for a TT-earthedinstallation is assured by the use of a RCD ofsensitivity:

I∆n i UL = 50 V RA RA

whereRA = the resistance of the earth electrode forthe installation.I∆n = rated differential current operating level.For temporary supplies (to work-sites etc.)and agricultural and horticulturalestablishments, the value of UL in the above-mentioned formula must be replaced by 25 V.

exampleThe resistance of the substation neutral earthelectrode Rn is 10 ohms.The resistance of the installation earthelectrode RA is 20 ohms.The earth-fault current Id = 7.7 A.The touch-voltage Uc = IdRA = 154 V andtherefore dangerous, but,

I∆n = 50 = 2.5 A so that a standard 300 mA 20RCD will operate in 30 ms to clear a conditionin which 50 V touch voltage, or more,appears on an exposed conductive part.

HV/400V

substation earth electrode

Rn : 10 Ω

installationearth electrode

RA : 20 Ω

1

2

3

4

Uc

fig. G10: automatic disconnection for a TT-earthed installation.

G



3.2 automatic disconnection for a TT-earthed installation (continued)

the tripping times of RCDs aregenerally lower than thoseprescribed in the majority of nationalstandards; this feature facilitates theiruse and allows the adoption of aneffective scheme of discriminativeprotection.

specified disconnection timeRCD is a general term for all devicesoperating on the residual-current principle.RCCB* (residual current circuit breaker) asdefined in IEC 1008 is a specific class ofRCD.Type G (general) and type S (selective) havetripping time/current characteristics as shownin table G11. These characteristics allow acertain degree of selective tripping betweenthe several combinations of rating and type,as shown later in Sub-clause 4.3.

x I∆n 1 2 5 > 5

instantaneous (ms) 300 150 40 40domestic

type S (ms) 500 200 150 150

industrialsetting I** (ms) 150 150 150 150

* Merlin Gerin

table G11: maximum operating times ofRCCBs (IEC 1008).** Note : the use of the term "circuit breaker" does not meanthat a RCCB can break short-circuit currents. For suchduties RCDs known as RCBOs ("O" for overcurrent) asdefined in IEC 1009 must be employed.

3.3 automatic disconnection for a TN-earthed installation

the principle of the TN scheme ofearthing is to ensure that earth-faultcurrent will be sufficient to operateovercurrent protective devices(direct-acting tripping, overcurrentrelays and fuses) so that

Ia i Uo or 0.8 Uo* Zs Zc

principleIn this scheme all exposed and extraneousconductive parts of the installation areconnected directly to the earthed point of thepower supply by protective conductors.As noted in Chapter F Sub-clause 4.2, theway in which this direct connection is carriedout depends on whether the TN-C, TN-S, orTN-C-S method of implementing the TNprinciple is used. In figure G12 the methodTN-C is shown, in which the neutralconductor acts as both the Protective-Earthand Neutral (PEN) conductor. In all TNarrangements, any insulation fault to earthconstitutes a phase-neutral short-circuit.High fault current levels simplify protectionrequirements but can give rise to touchvoltages exceeding 50% of the phase-to-neutral voltage at the fault position during thebrief disconnection time.In practice, therefore, earth electrodes arenormally installed at intervals along theneutral of the supply network, while theconsumer is generally required to instal anearth electrode at the service position. Onlarge installations additional earth electrodesdispersed around the premises are often

provided, in order to reduce the touch voltageas much as possible.In high-rise apartment blocks, all extraneousconductive parts are connected to theprotective conductor at each level.In order to ensure adequate protection, theearth-fault current

Id = Uo or 0.8 Uo u Ia where Zs ZcUo = nominal phase-neutral voltage.Zs = earth-fault current loop impedance,equal to the sum of the impedances of: thesource, the live phase conductors to the faultposition, the protective conductors from thefault position back to the source.Zc = the faulty-circuit loop impedance (see"conventional method" Sub-clause 5.2).Note: the path through earth electrodes backto the source will have (generally) muchhigher impedance values than those listedabove, and need not be considered.Id = the fault current.Ia = a current equal to the value required tooperate the protective device in the timespecified.

example

RnA

N

321

PEN

A

F

NS160

50 m35 mm2

CD

35mm2

E

B

Uc

fig. G12: automatic disconnection for a TN-earthed installation.

In figure G12 the touch voltage

Uc = 230 = 115 V 2and is therefore dangerous.The impedance Zs of the loop = ZAB + ZBC +ZDE + ZEN + ZNA.If ZBC and ZDE are predominant, then:

ZS = 2ρ x L = 64.3 milli-ohm, so that SId = 230/64.3 = 3,576 A (≈ 22 In based on a160 A circuit breaker).The "instantaneous" magnetic tripping devicesetting of the circuit breaker is many times

less than this value, so that positive operationin the shortest possible time is assured.Note: some authorities base suchcalculations on the assumption that a voltagedrop of 20% occurs in the part of theimpedance loop BANE.This method, which is recommended, isexplained in chapter G Sub-clause 5.2"conventional method" and, in this example,will give a fault current of230 x 0.8 x 103 = 2,816 A (≈ 18 In) 64.3

G


for TN earthing, the maximumallowable disconnection timedepends on the nominal voltage ofthe system.

specified maximumdisconnection timesThe times specified are a function of thenominal voltage phase/earth, which, for allpractical purposes on TN systems, is thephase/neutral voltage.

Uo (volts) disconnection timephase/neutral (seconds) U L=50 V

(see note 2)127 0.8230 0.4400 0.2> 400 0.1

table G13: maximum disconnection timesspecified for TN earthing schemes(IEC 364-4-41).

certain national regulations impose, theprovision of equipotential bonding of allextraneous and exposed conductive partsthat are simultaneously accessible, in anyarea where socket-outlets are installed, fromwhich portable or mobile equipment might besupplied. The common equipotential busbaris installed in the distribution-board cabinetfor the area concerned.

Note 2: when the conventional voltage limit is25 V, the specified disconnection times are:0.35 s. for 127 V0.2 s. for 230 V0.05 s. for 400 VIf the circuits concerned are final circuits, thenthese times can easily be achieved by theuse of RCDs.Note 3: the use of RCDs may, as mentionedin note 2, be necessary on TN-earthedsystems. Use of RCDs on TN-C-S systemsmeans that the protective conductor and theneutral conductor must (evidently) beseparated upstream of the RCD. Thisseparation is commonly made at the serviceposition.

Note 1: a longer time interval than thosespecified in the table (but in any case lessthan 5 seconds) is allowed under certaincircumstances for distribution circuits, as wellas for final circuits supplying a fixedappliance, on condition that a dangeroustouch voltage is not thereby caused to appearon another appliance. IEC recommends, and

protection by means of a circuit

breaker

The instantaneous trip unit of a circuit breakerwill eliminate a short-circuit to earth fault inless than 0.1 second.In consequence, automatic disconnectionwithin the maximum allowable time willalways be assured, since all types of trip unit,magnetic or electronic, instantaneous orslightly retarded, are suitable: Ia = Im.The maximum tolerance authorized by therelevant standard, however, must always betaken into consideration.It is sufficient therefore that the fault currentUo / Zs or 0.8 Uo / Zc determined bycalculation (or established on site) be greaterthan the instantaneous trip-setting current, orthan the very short-time tripping thresholdlevel, to be sure of tripping whithin thepermitted time limit.

if the protection is to be provided by acircuit breaker, it is sufficient to verifythat the fault current will alwaysexceed the current-setting level ofthe instantaneous or short-time delaytripping unit (Im):

Im < Uo or 0.8*Zs Zc

* according to the "conventional" method ofcalculation (see sub-clause 5.2).

1

Im Uo/Zs

t

I

2

1 : instantaneous trip2 : short time-delayed trip

fig. G14: disconnection by circuit breakerfor a TN-earthed installation.

protection by means of fusesThe value of current which assures thecorrect operation of a fuse can beaccertained from a current/time performancegraph for the fuse concerned.The fault current Uo/Zs or 0.8 Uo/Zc asdetermined above, must largely exceed thatnecessary to ensure positive operation of thefuse.The condition to observe therefore is that:

Ia < Uo or 0.8 Uo as indicated in figure G15. Zs Zc

Example:The nominal phase-neutral voltage of thenetwork is 230 V and the maximumdisconnection time given by the graph infigure G15 is 0.4 seconds. The correspondingvalue of Ia can be read from the graph.Using the voltage (230 V) and the current Ia,the complete loop impedance or the circuitloop impedance can be calculated fromZs = 230/Ia or Zc = 0.8 x 230/Ia.Thisimpedance value must never be exceededand should preferably be substantially less toensure satisfactory fuse operation.

I

t

tc

Ia

= 0,4 s

Uo/Zs

fig. G15: disconnection by fusesfor a TN-earthed installation.

Ia can be determined from the fuseperformance curve. In any case,protection cannot be achieved if theloop impedance Zs or Zc exceeds acertain value.

G


3.4 automatic disconnection on a second earth fault in an IT-earthed system


In this type of system:c the installation is isolated from earth, or theneutral point of its power-supply source isconnected to earth through a highimpedance,

c all exposed and extraneous conductiveparts are earthed via an installation earthelectrode.

in an IT scheme it is intended that afirst fault to earth will not cause anydisconnection.

first faultOn the occurrence of a short-circuit fault toearth, referred to as a "first fault", the faultcurrent is very small, such that the ruleId x RA i 50 V (see G3.2)is respected and no dangerous touchvoltages can occur.In practice the current Id is feeble, a conditionthat is neither dangerous to personnel, norharmful to the installation.However, in this scheme:c a permanent surveillance of the condition ofthe insulation to earth must be provided,together with an alarm signal (audio and/orflashing lights, etc.) in the event of a firstearth fault occurring,c the rapid location and repair of a first fault isimperative if the full benefits of the IT systemare to be realized. Continuity of service is thegreat advantage afforded by the scheme.

fig. G16: phases to earth insulationmonitoring relay (obligatory on IT-earthedinstallation).

RnA = 5 Ω

321

PE

HV/400 V

A

B

Zct1500 Ω

Id1

ZF

Id2

Uc

Id1Id2

Id2

Id2

Id1

Id2

Id1

fig. G17: fault-current paths for a first (earth) fault on an IT-earthed installation.

Example:For a network formed from 1 km ofconductors, the leakage (capacitive)impedance to earth ZF is of the order of3,500 ohms per phase. In normal (unfaulted)operation, the capacitive current* to earth isthereforeUo = 230 = 66 mA per phaseZF 3,500During a phase-to-earth fault, as shown infigure G17, the current passing through theelectrode resistance RnA is the vector sum ofthe capacitive currents in the two healthyphases. The voltages of the two healthyphases have (because of the fault) increasedto √3 the normal phase voltage, so that thecapacitive currents increase by the sameamount. These currents are displaced, onefrom the other by 60°, so that when addedvectorially, this amounts to 3 x 66 mA =198 mA i.e. Id2 in the present example.

The touch voltage Uc is therefore198 x 5 x 10-3 = 0.99 V, which is evidentlyharmless.The current through the short-circuit is givenby the vector sum of the neutral-resistorcurrent Id1 (= 153 mA) and the capacitivecurrent (Id2).Since the exposed conductive parts of theinstallation are connected directly to earth,the neutral impedance Zct plays practically nopart in the production of touch voltages toearth.* Resistive leakage current to earth through the insulation isassumed to be negligibly small in the example.

second fault situationOn the appearance of a second fault, on adifferent phase, or on a neutral conductor,a rapid disconnection becomes imperative.Fault clearance is carried out differently ineach of the following cases:1st case: concerns an installation in which allexposed conductive parts are bonded to acommon PE conductor, as shown in figureG19.In this case no earth electrodes are includedin the fault current path, so that a high level of

fault current is assured, and conventionalovercurrent protective devices are used, i.e.circuit breakers and fuses.The first fault could occur at the end of acircuit in a remote part of the installation,while the second fault could feasibly belocated at the opposite end of the installation.For this reason, it is conventional to doublethe loop impedance of a circuit, whencalculating the anticipated fault setting levelfor its overcurrent protective device(s).

the simultaneous existence of twoearth faults (if not both on the samephase) is dangerous, and rapidclearance by fuses or automaticcircuit breaker tripping depends onthe type of earth-bonding scheme,and whether separate earthingelectrodes are used or not, in theinstallation concerned.

G


c if no neutral conductor is provided, then thevoltage to use for the fault-current calculationis the phase-to-phase value, i.e.

(2) 0.8 e Uo* u Ia 2 ZcSpecified tripping/fuse-clearance timesDisconnecting times for 3-wire 3-phase ITschemes differ from those adopted for 4-wire3-phase IT schemes, and are given for bothcases in table G18.* based on the "conventional method" noted in the firstexample of Sub-clause 3.3.

1st case: where all exposedconductive parts are connected to acommon PE conductor conventionalovercurrent protection schemes(such as those used in TN systems)are applicable, with fault-levelcalculations and tripping/fuse-operating times suitably adapted.

c where the system includes a neutralconductor in addition to the 3 phaseconductors, the lowest short-circuit faultcurrents will occur if one of the (two) faults isfrom the neutral conductor to earth (all fourconductors are insulated from earth in an ITscheme). In four-wire IT installations,therefore, the phase-to-neutral voltage mustbe used to calculate short-circuit protectivelevels i.e.

(1) 0.8 Uo* u Ia where 2 ZcUo = phase/neutral voltageZc = impedance of the circuit fault-currentloop (see G3.3)Ia = current level for trip setting

Uo/U (volts) disconnection time (seconds) U L = 50 V (1)

Uo = phase-neutral volts 3-phase 3-wires 3-phase 4-wiresU = phase-phase volts127/220 0.8 5230/400 0.4 0.8400/690 0.2 0.4580/1000 0.1 0.2

table G18: maximum disconnection times specified for an IT-earthed installation(IEC 364-4-41).

(1) When the conventional voltage limit is25 V, the disconnecting times become:c in the case of a 3-phase 3-wire scheme,0.4 seconds at 127/220 V; 0.2 seconds at230/400 V and 0.06 seconds at 400/690 V,

c in the case of a 3-phase 4-wire scheme,1.0 second at 127/220 V; 0.5 seconds at230/400 V and 0.2 seconds at 400/690 V.

Example

Rn

321

PE

HV/400 V

HZct

IdA

K

RA

G D

E

C

50 m35 mm2

NS160160 A

F

B

J

busbars

50 m35 mm2

fig. G19: circuit breaker tripping on second (earth) fault when exposed conductive partsare connected to a common protective conductor.

The current levels and protective measuresdepend on the switchgear and fusegearconcerned:ccccc circuit breakersIn the case shown in figure G19, the levels ofinstantaneous and short time-delayovercurrent-trip settings must be decided.The times recommended in table G18 can bereadily complied with.Example: from the case shown in figure G19,determine that the short-circuit protectionprovided by the 160 A circuit breaker issuitable to clear a phase-to-phase short-circuit occurring at the load ends of thecircuits concerned.Reminder: In an IT system, the two circuitsinvolved in a phase-to-phase short circuit areassumed to be of equal length, with the samesized conductors; the PE conductors beingthe same size as the phase conductors.In such a case, the impedance of the circuitloop when using the "conventional method"(Sub-clause 5.2 of this chapter) will be twicethat calculated for one of the circuits in theTN case, shown in Sub-clause 3.3.

So that the resistance of circuit 1 loop FGHJ

= 2 RHJ = 2 ρ l mΩ awhere: ρ = the resistance in milli-ohm of acopper rod 1 metre long of c.s.a. 1 mm2

l = length of the circuit in metresa = c.s.a. of the conductor in mm2

= 2 x 22.5 x 50 = 64.3 mΩ 35and the loop resistance B, C, D, E, F, G, H, Jwill be 2 x 64.3 = 129 mΩThe fault current will therefore be:0.8 x ex 230 x 103 = 2.470 A 129c fusesThe current Ia for which fuse operation mustbe assured in a time specified according totable G18 can be found from fuse operatingcurves, as described in figure G15.The current indicated should be significantlylower than the fault currents calculated for thecircuit concerned,c RCCBsIn particular cases, RCCBs are necessary. Inthis case, protection against indirect contacthazards can be achieved by using one RCCBfor each circuit.

G


3.4 automatic disconnection on a second earth fault in an IT-earthed system (continued)


2nd case: where exposed conductiveparts of appliances are earthedindividually or in separate groups,each appliance or each group must(in addition to overcurrent protection)be protected by a RCD.

2nd case: concerns exposed conductiveparts which are earthed either individually(each part having its own earth electrode) orin separate groups (one electrode for eachgroup).If all exposed conductive parts are notbonded to a common electrode system, thenit is possible for the second earth fault tooccur in a different group or in a separately-earthed individual apparatus. Additionalprotection to that described above for case 1,is required, and consists of a RCD placed atthe circuit breaker controlling each group andeach individually-earthed apparatus.The reason for this requirement is that theseparate-group electrodes are "bonded"through the earth so that the phase-to-phaseshort-circuit current will generally be limitedwhen passing through the earth bond, by the

electrode contact resistances with the earth,thereby making protection by overcurrentdevices unreliable. The more sensitive RCDsare therefore necessary, but the operatingcurrent of the RCDs must evidently exceedthat which occurs for a first fault.For a second fault occurring within a grouphaving a common earth-electrode system,the overcurrent protection operates, asdescribed above for case 1.Note 1: see also Chapter H1 Sub-clause 7.2,protection of the neutral conductor.Note 2: in 3-phase 4-wire installationsprotection against overcurrent in the neutralconductor is sometimes more convenientlyachieved by using a ring-type currenttransformer over the single-core neutralconductor, as shown in figure G20 (see alsoTable H1-65c).

Rn

HV/LV

PIM

RA

RCD

groupearth

case 1

N

Rn

HV/LV

PIM

RA 2

RCD RCD

RCD

RA 1

group 1earth

group 2earth

N

case 2

fig. G20: the application of RCDs when exposed conductive parts are earthed individuallyor by groups, on IT-earthed systems.

3.5 measures of protection against direct or indirect contact without circuit disconnection

extra-low voltage is used where therisks are great: swimming pools,wandering-lead hand lamps, andother portable appliances for outdooruse, etc.

the use of SELV (Safety by Extra

Low Voltage)

Safety by extra low voltage SELV is used insituations where the operation of electricalequipment presents a serious hazard(swimming pools, amusement parks, etc.).This measure depends on supplying power atvery low voltage from the secondary windingsof isolating transformers especially designedaccording to national or to international(IEC 742) standards.The impulse withstand level of insulationbetween the primary and secondary windingsis very high, and/or an earthed metal screenis sometimes incorporated between thewindings. The secondary voltage neverexceeds 50 V rms.Three conditions of exploitation must berespected in order to provide satisfactoryprotection against indirect contact:c no live conductor at SELV must beconnected to earth,c exposed conductive parts of SELV-supplied

equipment must not be connected to earth, toother exposed conductive parts, or toextraneous conductive parts,c all live parts of SELV circuits and of othercircuits of higher voltage must be separatedby a distance at least equal to that betweenthe primary and secondary windings of asafety isolating transformer.These measures require that:c SELV circuits must use conduits exclusivelyprovided for them, unless cables which areinsulated for the highest voltage of the othercircuits are used for the SELV circuits,c socket outlets for the SELV system mustnot have an earth-pin contact. The SELVcircuit plugs and sockets must be special, sothat inadvertent connection to a differentvoltage level is not possible.Note: In normal conditions, when the SELVvoltage is less than 25 V, there is no need toprovide protection against direct-contacthazards. Particular requirements areindicated in Chapter L, Clause 3: "speciallocations".

the use of PELV (Protection by

Extra Low Voltage)

This system is for general use where lowvoltage is required, or preferred for safetyreasons, other than in the high-risk locationsnoted above. The conception is similar to thatof the SELV system, but the secondary circuitis earthed at one point.IEC 364-4-41 defines precisely thesignificance of the reference PELV. Protectionagainst direct-contact hazards is generallynecessary, except when the equipment is inthe zone of equipotential bonding, and thenominal voltage does not exceed 25 V rms,

and the equipment is used in normally drylocations only, and large-area contact with thehuman body is not expected.In all other cases, 6 V rms is the maximumpermitted voltage, where no direct-contactprotection is provided.

fig. G21: low-voltage supplies from a safetyisolating transformer, as defined in IEC 742.

230 V / 24 V

G


FELV system (Functional Extra

Low Voltage)

Where, for functional reasons, a voltage of50 V or less is used, but not all of therequirements relating to SELV or PELV arefulfilled, appropriate measures described inIEC 364-4-41 must be taken to ensureprotection against both direct and indirectcontact hazards, according to the locationand use of these circuits.

Note: Such conditions may, for example, beencountered when the circuit containsequipment (such as transformers, relays,remote-control switches, contactors)insufficiently insulated with respect to circuitsat higher voltages.

the separation of electric circuitsThe principle of separation of circuits(generally single-phase circuits) for safetypurposes is based on the followingreasoning.The two conductors from the unearthedsingle-phase secondary winding of aseparation transformer are insulated fromearth.If a direct contact is made with one conductor,a very small current only will flow into theperson making contact, through the earth andback to the other conductor, via the inherentcapacitance of that conductor with respect toearth. Since the conductor capacitance toearth is very small, the current is generallybelow the level of perception.As the length of circuit cable increases, thedirect contact current will progressivelyincrease to a point where a dangerouselectric shock will be experienced.Even if a short length of cable precludes anydanger from capacitive current, a low value ofinsulation resistance with respect to earth canresult in danger, since the current path is thenvia the person making contact, through theearth and back to the other conductorthrough the low conductor-to-earth insulationresistance.For these reasons, relatively short lengths ofwell-insulated cable are essential inseparation schemes.Transformers are specially designed for thisduty, with a high degree of insulation betweenprimary and secondary windings, or withequivalent protection, such as an earthedmetal screen between the windings.Construction of the transformer is to class IIinsulation standards.As indicated above, successful exploitation ofthe principle requires that:

the separation of electric circuits issuitable for relatively short cablelengths and high levels of insulationresistance. It is preferably used foran individual appliance.

c no conductor or exposed conductive part ofthe secondary circuit must be connected toearth,c the length of secondary cabling must belimited to avoid large capacitance values*,c a high insulation-resistance value must bemaintained for the cabling and appliances.These conditions generally limit theapplication of this safety measure to anindividual appliance.In the case where several appliances aresupplied from a separation transformer, it isnecessary to observe the followingrequirements:c the exposed conductive parts of allappliances must be connected together by aninsulated protective conductor, but notconnected to earth,c the socket outlets must be provided with anearth-pin connection. The earth-pinconnection is used in this case only to ensurethe interconnection (bonding) of all exposedconductive parts.In the case of a second fault, overcurrentprotection must provide automaticdisconnection in the same conditions asthose required for an IT scheme of powersystem earthing.* It is recommended in IEC 364-4-41 that the product of thenominal voltage of the circuit in volts and length in metres ofthe wiring system should not exceed 100 000, and that thelength of the wiring system should not exceed 500 m.

separationtransformer230 V / 230 Vclass II

fig. G22: safety supplies from a separationtransformer.

class II appliances

symbolThese appliances are also referred to ashaving "double insulation" since in class IIappliances a supplementary insulation isadded to the basic insulation. No conductiveparts of a class II appliance must beconnected to a protective conductor:c most portable or semi-fixed appliances,certain lamps, and some types of transformerare designed to have double insulation. It isimportant to take particular care in theexploitation of class II equipment and to verifyregularly and often that the class II standardis maintained (no broken outer envelope,etc.). Electronic devices, radio and televisionsets have safety levels equivalent to class II,but are not formally class II appliances,c supplementary insulation in an electricalinstallation (IEC 364-4-41: Sub-clause 413-2).Some national standards such asNF C 15-100 (France) (annex to 413.5

active part

basicinsulationsupplementaryinsulation

fig. G23: principle of class II insulationlevel.

Chapter 41) describe in more detail thenecessary measures to achieve thesupplementary insulation during installationwork.A simple example is that of drawing a cableinto a PVC conduit. Methods are alsodescribed for distribution boards.c for distribution boards and similarequipment, IEC 439-1 describes a set ofrequirements, for what is referred to as "totalinsulation", equivalent to class II,c some cables are recognized as beingequivalent to class II by many nationalstandards.

G



3.5 measures of protection against direct or indirect contact without circuit disconnection(continued)

in principle, safety by placingsimultaneously-accessibleconductive parts out-of-reach, or byinterposing obstacles, requires also anon-conducting floor, and so is not aneasily applied principle

out-of-reach or interposition

of obstacles.

By these means, the probability of touching alive exposed conductive part, while at thesame time touching an extraneousconductive part at earth potential, isextremely low. In practice, this measure canonly be applied in a dry location, and isimplemented according to the followingconditions:c the floor and the walls of the chamber mustbe non-conducting, i.e. the resistance to earthat any point must be:> 50 kΩ (installation voltages i 500 V),> 100 kΩ (500 V < installation voltagesi 1000 V).Resistance is measured by means of"MEGGER" type instruments (hand-operatedgenerator or battery-operated electronicmodel) between an electrode placed on thefloor or against the wall, and earth (i.e. thenearest protective earth-conductor).The electrode contact area and pressure

must evidently be the same for all tests.Different instrument suppliers provideelectrodes specific to their own product, sothat care should be taken to ensure that theelectrodes used are those supplied with theinstrument. There are no universallyrecognized standards established for thesetests at the time of writing.c the placing of equipment and obstaclesmust be such that simultaneous contact withtwo exposed conductive parts or with anexposed conductive part and an extraneousconductive part by an individual person is notpossible.c no exposed protective conductor must beintroduced into the chamber concerned.c entrances to the chamber must bearranged so that persons entering are not atrisk, e.g. a person standing on a conductingfloor outside the chamber must not be able toreach through the doorway to touch anexposed conductive part, such as a lightingswitch mounted in an industrial-type cast-ironconduit box, for example.

electricalapparatus

electricalapparatus

< 2 m

electricalapparatus

insulatedwallsinsulated

obstacles

> 2 m

insulated floor

2.5 m

fig. G24: protection by out-of-reach arrangements and the interposition of non-conductingobstacles.

between a live conductor and the metalenvelope of an appliance will result in thewhole "cage" being raised to phase-to-earthvoltage, but no fault current will flow. In suchconditions, a person entering the chamberwould be at risk (since he/she would bestepping on to a live floor). Suitableprecautions must be taken to protectpersonnel from this danger (e.g. non-conducting floor at entrances, etc.).Special protective devices are also necessaryto detect insulation failure, in the absence ofsignificant fault current.*Note: extraneous conductive parts entering (or leaving) theequipotential space (such as water pipes, etc.) must beencased in suitable insulating material and excluded fromthe equipotential network, since such parts are likely to bebonded to protective (earthed) conductors elsewhere in theinstallation.

earth-free equipotential

chambers

In this scheme, all exposed conductive parts,including the floor (see *Note) are bonded bysuitably large conductors, such that nosignificant difference of potential can existbetween any two points. A failure of insulation

earth-free equipotential chambersare associated with particularinstallations (laboratories, etc.) andgive rise to a number of practicalinstallation difficulties.

insulatingmaterial

conductive floor

M

fig. G25: equipotential bonding of all exposed conductive partssimultaneously accessible.

G


4. implementation of the TT system

4.1 protective measures

the application to living quarters iscovered in Chapter L Clause 1.

protection against indirect

contact

General caseProtection against indirect contact is assuredby RCDs, the sensitivity I∆n of whichcomplies with the condition:

I∆n i 50 V (1)

RA(1) 25 V for work-site installations, agriculturalestablishments, etc.

The choice of sensitivity of the differentialdevice is a function of the resistance RA ofthe earth electrode for the installation, and isgiven in table G26.

I∆n maximum resistanceof the earth electrode(50 V) (25 V)

3 A 16 Ω 8 Ω1 A 50 Ω 25 Ω500 mA 100 Ω 50 Ω300 mA 166 Ω 83 Ω30 mA 1666 Ω 833 Ωtable G26: the upper limit of resistance foran installation earthing electrode whichmust not be exceeded, for givensensitivity levels of RCDs at U L voltagelimits of 50 V and 25 V.

Case of distribution circuitsIEC 364-4-41 and a number of nationalstandards recognize a maximum tripping timeof 1 second in installation distribution circuits(as opposed to final circuits). This allows adegree of selective discrimination to beachieved:c at level A: RCD time-delayed, e.g. "S" type,c at level B: RCD instantaneous.

A

B

fig. G27: distribution circuits.

Case where the exposed conductive partsof an appliance, or group of appliances,are connected to a separate earthelectrodeProtection against indirect contact by a RCDat the circuit breaker controlling each groupor separately-earthed individual appliance.In each case, the sensitivity must becompatible with the resistance of the earthelectrode concerned. RA 2RA 1

a distant location

fig. G28: separate earth electrode.

high-sensitivity RCDs

IEC 364-4-471 strongly recommends the useof a RCD of high sensitivity (i 30 mA) in thefollowing cases:c socket-outlet circuits for rated currentsof i 32 A at any location(1),c socket-outlet circuits in wet locations at allcurrent ratings(1),c socket-outlet circuits in temporaryinstallations(1),c circuits supplying laundry rooms andswimming pools(1),c supply circuits to work-sites, caravans,pleasure boats, and travelling fairs(1).This protection may be for individual circuitsor for groups of circuits,c strongly recommended for circuits of socketoutlets u 20 A (mandatory if they areexpected to supply portable equipment foroutdoor use),c in some countries, this requirement ismandatory for all socket-outlet circuitsrated i 32 A.(1) these cases are treated in delail in Chapter L Clause 3.

fig. G29: circuit supplying socket-outlets.

G


4. implementation of the TT system (continued)

4.1 protective measures (continued)

in areas of high fire risk

RCD protection at the circuit breakercontrolling all supplies to the area at risk isnecessary in some locations, and mandatoryin many countries.The sensitivity of the RCD must be i 500 mA.

fire-risk area

fig. G30: fire-risk location.

protection when exposed

conductive parts are not

connected to earth

(in the case of an existing installation wherethe location is dry and provision of anearthing connection is not possible, or in theevent that a protective earth wire becomesbroken)RCDs of high sensitivity (i 30 mA) will affordboth protection against indirect-contacthazards, and the additional protection againstthe dangers of direct-contact .

fig. G31: unearthed exposed conductiveparts (A).

4.2 types of RCD

RCDs are commonly incorporated in thefollowing components:c industrial-type moulded-case differentialcircuit breakers conforming to IEC 947-2 andits appendix B,c domestic-type differential circuit breakers(RCCBs)* conforming to IEC 755, 1008, and1009 (RCBOs),*see NOTE concerning RCCBs at the end ofSub-clause 3.2.

c differential switches conforming to particularnational standards,c relays with separate toroidal (ring-type)current transformers, conforming to IEC 755.RCDs are mandatorily used at the origin ofTT-earthed installations, where their ability todiscriminate with other RCDs allows selectivetripping, thereby ensuring the level of servicecontinuity required.

fig. G32: industrial-type CB with RCD module.DIN-rail circuit breaker with RCD module

Adaptable differential circuit breakers,including DIN-rail mounted units, areavailable, to which may be associated anauxiliary module. The ensemble provides a

comprehensive range of protective functions(isolation, short-circuit, overload, andsensitive earth-fault protection).

the international standard forindustrial differential circuit breakersis IEC 947-2 and its appendix B.

G


the international standards fordomestic circuit breakers (RCBOs) isIEC 1009.

The incoming-supply circuit breaker can also have time-delayed characteristics (type S).

"Monobloc" type of earth-fault differential circuit breakersdesigned for the protection of socket-outlet circuits and finalcircuit protection.

fig. G33: domestic earth-fault differential circuit breakers.

In addition to the adaptable industrial circuitbreakers which comply to industrial anddomestic standards, there are ranges of

"monobloc" differential circuit breakersintended for domestic and tertiary sectorapplications.

Differential switches (RCCBs) are used for the protection ofdistribution or sub-distribution boards.

RCDs with separate toroidal CTs can be used in associationwith circuit breakers or contactors.

fig. G34: differential switches (RCCBs). fig. G35: RCDs with separate toroidalcurrent transformers.

differential switches are covered byparticular national standards(NF C 61-140 for France).RCDs with separate toroidal currenttransformers are standardizedin IEC 755.

RCCBs, RCBOs and CBRs

RCCBs (Residual Current Circuit Breakers)These devices are more-accurately describedin the French version of IEC 1008 as"interrupteurs" which is generally translatedinto English by "load-break switches"."Residual-current load-break switches" wouldbe a more accurate description of a RCCB,which, although assigned a rated making andbreaking capacity, is not designed to breakshort-circuit currents (the unique feature of acircuit breaker) so that the term RCCB can bemisleading.As noted in sub-clause 7.3 a SCPD (Short-Circuit Protective Device) must always beseries-connected with a RCCB.

RCBOsThe "O" stands for "Overcurrent" which refersto the fact that, in addition to sensitivedifferential earth-fault protection, overcurrentprotection is provided. The RCBO has a ratedshort-circuit breaking capability and isproperly referred to as a circuit breaker.IEC 1009 is the international referencestandard.

Note: Both RCCBs and RCBOs asstandardized in IEC 1008 and 1009respectively, provide complete isolation whenopened. These units are designed fordomestic and similar installations.

CBRsAmendment 1 (1992) of the product standardIEC 947-Part 2: "Circuit Breakers" includesAppendix B, which covers the incorporationof residual-current protection into industrial-type LV circuit breakers. The Appendix isbased on the relevant requirements of IEC755,IEC 1008 and IEC 1009. Circuit breakers soequipped are referred to as CBRs.

4.3 coordination of differential protective devices

Discriminative-tripping coordination isachieved either by time-delay or bysubdivision of circuits, which are thenprotected individually or by groups, or by acombination of both methods.Such discrimination avoids the tripping of anycircuit breaker, other than that immediatelyupstream of a fault positionc with equipment currently available,discrimination is possible at three or fourdifferent levels of distribution, viz:v at the main general distribution board,

v at local general distribution boards,v at sub-distribution boards,v at socket outlets for individual applianceprotectionc in general, at distribution boards (and sub-distribution boards, if existing) and onindividual-appliance protection, devices forautomatic disconnection in the event of anindirect-contact hazard occurring are installedtogether with additional protection againstdirect-contact hazards.

G


A

B

RCD 300 mAtype S

RCD30 mA

4. implementation of the TT system (continued)

4.3 coordination of differential protective devices (continued)

discrimination between RCDsDiscrimination is achieved by exploiting theseveral levels of standardized sensitivity:30 mA, 100 mA, 300 mA and 1 A and thecorresponding tripping times, as shown belowin figure G36.

time(ms)

40

10

60

100130150200250

500

1000

300

10000

15 current(mA)

30

100

15060

300

500

600

1000

1 1,5 10 100 500 1000 (A)

300 mA selective RCDs(i.e. time-delayed)

domestic Stime delayed

RCD 30 mA general domestic

industrial(settings I and II )

I

II

and industrial setting 0

fig. G36: discrimination between RCDs.

discrimination at 2 levelsProtection:c level A: RCD time-delayed setting 1 (forindustrial device) type S (for domestic device)for protection against indirect contacts,c level B: RCD instantaneous, with highsensitivity on circuits supplying socket-outletsor appliances at high risk (washing machines,etc. See also Chapter L Clause 3).

fig. G37.

discrimination at 3 or 4 levelsProtection:c level A: RCD time-delayed (setting III),c level B: RCD time-delayed (setting II),c level C: RCD time-delayed (setting I) ortype S,c level D: RCD instantaneous.

A relay with separatetoroidal CT 3 Adelay time 500 ms

RCCB 1 Adelay time 250 ms

B

C

D

RCCB 300 mAdelay time 50 ms or type S

RCCB 30 mA

fig. G38: discrimination at 3 or 4 levels.

G


discriminative protection at three levels

Rp

Rn

RA

321NPE

N123PE

NPhPE

M

M

MERLIN GERIN

differential relaywith separate toroidalCT setting level ≤ 50/RAtime-delay setting level II

main circuit breaker

HV/LV

MERLIN GERIN

MERLIN GERIN

SM20

IN OUT

earth leakagecurrent monitor

differentialrelay withseparate CT

discontactor

T

VigicompactNS100setting level300 mA

T

TT E S T

30mA

XC40diff.30 mARCD

DPN Vigi30 mA

NS400 NS80H-MA

instantaneous300 mA

distributionbox

discontactor

RCDMCBMCB

MCB + RCD

300 mA type Stime-delayedRCD

remotely-controlledactuator

fig. G39: typical 3-level installation, showing the protection of distribution circuits in a TT-earthed system.One motor is provided with specific protection.

G


5. implementation of the TN system

5.1 preliminary conditions

At the design stage, the maximum permittedlengths of cable downstream of a controllingcircuit breaker (or set of fuses) must becalculated, while during the installation workcertain rules must be fully respected.

imposed conditionsCertain conditions must be observed, aslisted below and illustrated in figure G40.1. earth electrodes should be provided atevenly-spaced points (as far as practicalconditions allow) along the PE conductor.Note : This is not normally done for a singledomestic installation; one earth electrode onlyis usually required at the service position.2. the PE conductor must not pass throughferro-magnetic conduit, ducts, etc. or bemounted on steel work, since inductive and/or proximity effects can increase the effectiveimpedance of the conductor.3. in the case of a PEN conductor (a neutralconductor which is also used as a protectiveconductor), connection must be made directlyto the earth terminal of an appliance (see 3 infigure G40) before being looped to the neutralterminal of the appliance.4. where the conductor i 6 mm2 for copper or10 mm2 for aluminium, or where a cable ismovable, the neutral and protectiveconductors should be separated (i.e. aTN-S scheme should be adopted within theinstallation).5. earth faults should be cleared byovercurrent-protection devices, i.e. by fusesand circuit breakers.The foregoing list indicates the conditions tobe respected in the implementation of aTN scheme for the protection against indirectcontacts.

RpnA

PEN

TN-C

PE

TN-C-S

N

2 2

34

5

5 5

1

fig. G40: implementation of the TN systemof earthing.

note(1) the TN scheme requires that the LV neutral of the HV/LVtransformer, the exposed conductive parts of the substationand of the installation, and the extraneous conductive partsin the sub-station and installation, all be earthed to acommon earthing system.(2) for a substation in which the metering is at low-voltage, ameans of isolation is required at the origin of the LVinstallation, and the isolation must be clearly visible.(3) a PEN conductor must never be interrupted under anycircumstances. Control and protective switchgear for theseveral TN arrangements will be:c 3-pole when the circuit includes a PEN conductor,c preferably 4-pole (3 phases + neutral) when the circuitincludes a neutral with a separate PE conductor.

5.2 protection against indirect contact

Three methods of calculation arecommonly used:c the method of impedances, basedon the trigonometric addition of thesystem resistances and inductivereactances.c the method of composition.c the conventional method, based onan assumed voltage drop and theuse of prepared tables.

methods of determining levels of

short-circuit current

In TN-earthed systems, a short-circuit toearth will, in principle, always providesufficient current to operate an overcurrentdevice. The source and supply mainsimpedances are much lower that those of theinstallation circuits, so that any restriction inthe magnitude of earth-fault currents will bemainly caused by the installation conductors(long flexible leads to appliances greatlyincrease the "fault-loop" impedance, with acorresponding reduction of short-circuitcurrent).The most recent IEC recommendations forindirect-contact protection on TN earthingschemes only relates maximum allowabletripping times to the nominal system voltage(see table G13 in Sub-clause 3.3).The reasoning behind theserecommendations is that, for TN systems, thecurrent which must pass in order to raise thepotential of an exposed conductive part to50 V or more is so high that one of twopossibilities will occur:c either the fault path will blow itself clear,practically instantaneously, orc the conductor will weld itself into a solidfault and provide adequate current to operateovercurrent devices.To ensure correct operation of overcurrentdevices in the latter case, a reasonablyaccurate assessment of short-circuit earth-

fault current levels must be determined at thedesign stage of a project.A rigorous analysis requires the use of phase-sequence-component techniques applied toevery circuit in turn. The principle isstraightforward, but the amount ofcomputation is not considered justifiable,especially since the zero-phase-sequenceimpedances are extremely difficult todetermine with any resonable degree ofaccuracy in an average LV installation.Other simpler methods of adequate accuracyare preferred. Three practical methods are:c the "method of impedances", based on thesummation of all the impedances (positive-phase-sequence only) around the fault loop,for each circuit,c the "method of composition", which is anestimation of short-circuit current at theremote end of a loop, when the short-circuitcurrent level at the near end of the loop isknown,c the "conventional method" of calculating theminimum levels of earth-fault currents,together with the use of tables of values forobtaining rapid results.These methods are only reliable for the casein which the cables that make up the earth-fault-current loop are in close proximity (toeach other) and not separated by ferro-magnetic materials.

G


for calculations, modern practice is touse software agreed by NationalAuthorities, and based on themethod of impedances, such asECODIAL 2 (Merlin Gerin).National Authorities generally alsopublish Guides, which include typicalvalues, conductor lengths, etc.

method of impedancesThis method summates the positive-sequence impedances of each item (cable,PE conductor, transformer, etc.) included inthe earth-fault loop circuit from which theshort-circuit earth-fault current is calculated,using the formula:

where (∑R)2 = (the sum of all resistances inthe loop)2

I =U/ (∑R)2+(∑X)

2

and (∑X)2 = (the sum of all inductivereactances in the loop)2

and U = nominal system phase-to-neutralvoltageThe application of the method is not alwayseasy, because it supposes a knowledge of allparameter values and characteristics of theelements in the loop. In many cases, anational guide can supply typical values forestimation purposes.

method of compositionThis method permits the determination of theshort-circuit current at the end of a loop fromthe known value of S.C. at the sending end,by means of the approximate formula:

I = U Isc where U + Zsc IscIsc = upstream short-circuit currentI = end-of-loop short-circuit currentU = nominal system phase voltageZsc = impedance of loop

Note: in this method the individualimpedances are added arithmetically* asopposed to the previous "method ofimpedances" procedure.

* This results in a calculated current value which is less thanthat which would actually flow. If the overcurrent settings arebased on this calculated value, then operation of the relay,or fuse, is assured.

conventional methodThis method is generally considered to besufficiently accurate to fix the upper limit ofcable lengths.Principle:The principle bases the short-circuit currentcalculation on the assumption that thevoltage at the origin of the circuit concerned(i.e. at the point at which the circuit protectivedevice is located) remains at 80% or more ofthe nominal phase to neutral voltage. The80% value is used, together with the circuitloop impedance, to compute the short-circuitcurrent.This coefficient takes account of all voltagedrops upstream of the point considered. In LVcables, when all conductors of a 3-phase4-wire circuit are in close proximity (which isthe normal case), the inductive reactanceinternal to* and between conductors isnegligibly small compared to the cableresistance.

This approximation is considered to be validfor cable sizes up to 120 mm2.Above that size, the resistance value R isincreased as follows:core size (mm2) value of resistanceS = 150 mm2 R+15%S = 185 mm2 R+20%S = 240 mm2 R+25%

* causes proximity and skin effects, i.e. an apparentincrease in resistance.

Example:

Id

L

C

PE

SPE Sph

AB

the maximum length of any circuit ofa TN-earthed installation is:

Lmax = 0.8 Uo Sph ρ (1+m) Ia

fig. G41: calculation of L max. for aTN-earthed system, using theconventional method.

The maximum length of a circuit in a TN-earthed installation is given by the formula:

Lmax = 0.8 Uo Sph metres, where: ρ (1+m) IaLmax = maximum length in metresUo = phase volts = 230 V for a 230/400 Vsystemρ = resistivity at normal working temperaturein ohm-mm2/metre= 22.5 10-3 for copper= 36 10-3 for aluminiumIa = trip current setting for the instantaneousoperation of a circuit breaker, orIa = the current which assures operation ofthe protective fuse concerned, in thespecified time.m = Sph / SPESph = cross-sectional area of the phaseconductors of the circuit concerned in mm2

SPE = cross-sectional area of the protectiveconductor concerned in mm2

G


nominal cross- instantaneous or short-time-delayed tripping current Im (amperes)sectional areaof conductorsmm 2 50 63 80 100 125 160 200 250 320 400 500 560 630 700 800 875 1000 1120 1250 1600 2000 2500 3200 4000 5000 6300 8000 10000 125001.5 103 81 64 51 41 32 25 20 16 13 10 9 8 7 6 6 52.5 171 136 107 85 66 53 42 34 26 21 17 15 13 12 10 10 8 8 7 54 274 217 171 137 109 85 68 54 43 34 27 24 21 19 17 16 14 12 11 8 7 56 410 326 256 205 164 126 102 82 64 51 41 36 32 29 25 23 20 18 16 13 10 8 6 510 427 342 273 214 171 137 107 85 68 61 54 49 42 39 34 30 27 21 17 14 10 8 7 516 436 342 274 219 171 137 109 97 87 78 68 62 55 49 44 34 27 21 17 13 11 8 7 525 428 342 267 213 171 152 135 122 107 98 85 76 66 53 43 34 27 21 17 13 10 8 735 479 374 299 239 214 190 171 150 136 120 107 96 75 80 48 37 30 24 19 15 12 950 406 325 290 258 232 203 185 162 145 130 101 81 65 50 40 32 26 20 16 1270 479 427 380 342 299 274 239 214 191 150 120 96 75 60 48 38 30 24 1995 464 406 371 325 290 260 203 162 130 101 81 65 51 40 32 26120 469 410 366 328 256 205 165 128 102 82 65 51 41 33150 446 398 357 279 223 178 139 111 89 71 56 44 36185 471 422 329 264 211 165 132 105 84 66 53 42240 410 328 263 205 164 131 104 82 66 52

5. implementation of the TN system (continued)

5.2 protection against indirect contact (continued)

the following tables* give the lengthof circuit which must not beexceeded, in order that persons beprotected against indirect contacthazards by protective devices.

* Based on tables given in the guide UTE C15-105.

tablesThe following tables, applicable to TNsystems, have been established according tothe "conventional method" described above.The tables give maximum circuit lengths,beyond which the ohmic resistance of theconductors will limit the magnitude of theshort-circuit current to a level below thatrequired to trip the circuit breaker (or to blowthe fuse) protecting the circuit, with sufficientrapidity to ensure safety against indirectcontact.

The tables take into account:c the type of protection: circuit breakers orfuses,c operating-current settings,c cross-sectional area of phase conductorsand protective conductors,c type of earthing scheme (see fig. G47),c type of circuit breaker (i.e. B, C or D).The tables may be used for 230/400 Vsystems.Equivalent tables for protection by Compactand Multi 9 circuit breakers (Merlin Gerin) areincluded in the relevant catalogues.

Correction factor mTable G42 indicates the correction factor toapply to the values given in tables G43 toG46 according to the ratio SPH/SPE, the typeof circuit, and the conductor materials.

circuit conductor m = S PH/SPE (or PEN)material m = 1 m = 2 m = 3 m = 4

3P + N or P + N copper 1 0.67 0.50 0.40aluminium 0.62 0.42 0.31 0.25

table G42: correction factor to apply to the lengths given in tables G43 to G46 forTN systems.

Circuits protected by general-purposecircuit-breakers

table G43: maximum circuit lengths for different sizes of conductor and instantaneous-tripping-current settings for general-pur posecircuit breakers.

Circuits protected by Compact* or Multi 9*circuit breakers for industrial or domesticuse

SPH rated current (A)mm 2 1 2 3 4 6 8 10 13 16 20 25 32 40 45 50 63 80 1001.5 1227 613 409 307 204 153 123 94 77 61 49 38 31 27 25 19 15 122.5 681 511 341 256 204 157 128 102 82 64 51 45 41 32 28 204 1090 818 545 409 327 252 204 164 131 102 82 73 65 52 41 336 818 613 491 377 307 245 196 153 123 109 98 78 61 4910 1022 818 629 511 409 327 256 204 182 164 130 102 8216 1006 818 654 523 409 327 291 262 208 164 13125 1022 818 639 511 454 409 325 258 20435 894 716 636 572 454 358 28850 777 617 485 388

table G44: maximum circuit lengths for different sizes of conductor and rated currents fortype B (1) circuit breakers.* Merlin Gerin products.(1) For the definition of type B circuit breaker refer to chapter H2 Sub-clause 4.2.

G


SPH rated current (A)mm2 1 2 3 4 6 8 10 13 16 20 25 32 40 45 50 63 80 1001.5 613 307 204 153 102 77 61 47 38 31 25 19 15 14 12 10 8 62.5 1022 511 341 256 170 128 102 79 64 51 41 32 26 23 20 16 13 104 818 545 409 273 204 164 126 102 82 65 51 41 36 33 26 20 166 818 613 409 307 245 189 153 123 98 77 61 55 49 39 31 2510 1022 681 511 409 315 256 204 164 128 102 91 82 65 51 4116 818 654 503 409 327 262 204 164 145 131 104 82 6525 1022 786 639 511 409 319 256 227 204 162 128 10235 894 716 572 447 358 318 286 227 179 14350 777 607 485 431 389 309 243 194

table G45: maximum circuit lengths for different conductor sizes and for rated currentsof circuit breakers of type C (1).(1) For the definition of type C circuit breakers refer to chapter H2 Sub-clause 4.2.

SPH rated current (A)mm2 1 1.6 2 2.5 3 4 6 6.3 8 10 12.5 13 16 20 25 32 40 45 50 63 80 100

1.5 438 274 219 175 146 110 73 70 53 44 35 34 27 22 18 14 11 10 9 7 5 4

2.5 730 456 365 292 243 183 122 116 88 73 58 56 46 37 29 23 18 16 15 12 9 7

4 730 584 467 389 292 195 186 141 117 93 90 73 58 47 37 29 26 23 19 14 12

6 876 701 584 438 292 279 211 175 140 135 110 88 70 55 44 39 35 28 21 18

10 974 730 487 465 352 292 234 225 183 146 117 91 73 65 58 46 35 29

16 779 743 564 467 374 359 292 234 187 146 117 104 93 74 58 47

25 881 730 584 562 456 365 292 228 183 162 146 116 88 73

35 1022 818 786 639 511 409 319 258 227 204 162 123 102

50 867 692 558 432 347 308 277 220 174 139

table G46: maximum circuit lengths for different conductor sizes and for rated currentsof circuit breakers of type D or MA Merlin Gerin (1).

(1) For the definition of type D circuit breakers refer to chapter H2 Sub-clause 4.2.For typical use of an MA circuit breaker, refer to Chapter J figure J5-3.

Example:A 3-phase 4-wire (230/400 V) installation isTN-C earthed. A circuit is protected by acircuit breaker rated at 63 A, and consists ofan aluminium cored cable with 50 mm2 phaseconductors and a neutral conductor (PEN) of25 mm2.What is the maximum length of circuit, belowwhich protection of persons against indirect-contact hazards is assured by theinstantaneous magnetic tripping relay of thecircuit breaker?Table G44 gives 617 metres, to which mustbe applied a factor of 0.42 (table G42 form = SPH/SPE = 2).The maximum length of circuit is therefore:617 x 0.42 = 259 metres.

particular case where one

or more exposed conductive

part(s) is (are) earthed to a

separate earth electrode

Protection must be provided against indirectcontact by a RCD at the origin of any circuitsupplying an appliance or group ofappliances, the exposed conductive parts ofwhich are connected to an independent earthelectrode.The sensitivity of the RCD must be adaptedto the earth electrode resistance (RA2 infigure G47).Downstream of the RCD, the earthingscheme must be TN-S.

RA 2RA 1

a distant location

fig. G47: separate earth electrode.

G



5. implementation of the TN system (continued)

5.3 high-sensitivity RCDs


5.4 protection in high fire-risk locations

In locations where the risk of fire is high, theTN-C scheme of earthing is often prohibited,and the TN-S arrangement must be adopted.Protection by a RCD of sensitivity 500 mA atthe origin of the circuit supplying the fire-risklocation is mandatory in some countries.

fire-risk area


G


5.5 when the fault-current-loop impedance is particularly high

When the earth-fault current is restricted dueto an inevitably high fault-loop impedance, sothat the overcurrent protection cannot berelied upon to trip the circuit within theprescribed time, the following possibilitiesshould be considered:

2 i Irm i 4In

PE or PEN

unusuallylong cable

fig. G50: a circuit breaker with low-setinstantaneous magnetic trip.

Suggestion 1:install a circuit breaker which has aninstantaneous magnetic tripping element withan operation level which is lower than theusual setting, for example:2In i Irm i 4InThis affords protection for persons on circuitswhich are abnormally long. It must bechecked, however, that high transientcurrents such as the starting currents ofmotors will not cause nuisance trip-outs.

Suggestion 2:install a RCD on the circuit. The device neednot be highly-sensitive (HS) (several amps toa few tens of amps). Where socket-outletsare involved, the particular circuits must, inany case, be protected by HS (i 30 mA)RCDs; generally one RCD for a number ofsocket outlets on a common circuit.

phasesneutralPE

TN-S

phasesPEN

TN-C

fig. G51: RCD protection on TN systemswith high earth-fault-loop impedance.

Suggestion 3:increase the size of the PE or PENconductors and/or the phase conductors, toreduce the loop impedance.

Suggestion 4:add supplementary equipotential conductors.This will have a similar effect to that ofsuggestion 3, i.e. a reduction in the earth-fault-loop resistance, while at the same timeimproving the existing touch-voltageprotection measures. The effectiveness ofthis improvement may be checked by aresistance test between each exposedconductive part and the local main protectiveconductor.For TN-C installations, bonding as shown infigure G52 is not allowed, and Suggestion 3should be adopted.

fig. G52: improved equipotential bonding.

G


6. implementation of the IT system

The basic feature of the IT scheme ofearthing is that, in the event of a short-circuitto earth fault, the system can continue tofunction without interruption.Such a fault is referred to as a "first fault". Inthis scheme, all exposed conductive parts ofan installation are connected via PEconductors to an earth electrode at theinstallation, while the neutral point of thesupply transformer is isolated from earth orconnected to earth through a high resistance(commonly 1,000 ohms or more).This means that the current through an earthfault will be measured in milli-amps, which willnot cause serious damage at the faultposition, or give rise to dangerous touchvoltages, or present a fire hazard. Thesystem may therefore be allowed to functionnormally until it is convenient to isolate thefaulty section for repair work.In practice, the scheme requires certainspecific measures for its satisfactoryexploitation:

c permanent monitoring of the insulation withrespect to earth, which must signal (audiblyor visually) the occurrence of the first fault,c a device for limiting the voltage which theneutral point of the supply transformer canattain with respect to earth,c a "first-fault" location routine by an efficientmaintenance staff. Fault location is greatlyfacilitated by automatic devices which arecurrently available,c automatic high-speed tripping ofappropriate circuit breakers must take placein the event of a "second fault" occurringbefore the first fault is repaired. The secondfault (by definition) is an earth fault affecting adifferent phase than that of the first fault or aneutral conductor*.The second fault results in a short-circuitthrough the earth and/or through PE bondingconductors.* on systems where the neutral is distributed, as shown infigure G58.

6.1 preliminary conditions

Preliminary conditions are summarized intable G53 and fig. G54.

minimum functions components examplesrequired and devices (MG)protection against overvoltages (1) voltage limiter Cardew Cat system frequencyneutral earthing resistor (2) resistor impedance Zx(for impedance earthing variation)overall earth-fault monitor (3) permanent insulation Vigilohm TR22Awith alarm for first fault condition monitor PIM with alarm feature or XM 200automatic fault clearance (4) four-pole circuit breakers Compact circuit breakeron second fault and (if the neutral is distributed) or RCD-MSprotection of the neutral all 4 poles + tripconductor against overcurrentlocation of first fault (5) with device for fault-location Vigilohm system

on live system, or by successiveopening of circuits

table G53: essential functions in IT schemes.

L1L2L3N

HV/LV

4

4

2 1 3

5

4

fig. G54: 3-phase 3-wire IT-earthedsystem.

G


6.2 protection against indirect contact

first-fault conditionThe earth-fault current which flows under afirst-fault condition is measured in milli-amps.The touch voltage with respect to earth is theproduct of this current and the resistance ofthe installation earth electrode and PEconductor (from the faulted component to theelectrode). This value of voltage is clearlyharmless and could amount to several voltsonly in the worst case (1,000 Ω earthingresistor will pass 230 mA* and a poorinstallation earth-electrode of 50 ohms, wouldgive 12.5 V, for example).An alarm is given by the permanent earth-fault monitoring.

Principle of earth-fault monitoringA generator of very low frequency a.c.current, or of d.c. current, (to reduce theeffects of cable capacitance to negligiblelevels) applies a voltage between the neutralpoint of the supply transformer and earth.This voltage causes a small current to flowaccording to the insulation resistance to earthof the whole installation, plus that of anyconnected appliance.

Low-frequency instruments can be used ona.c. systems which generate transient d.c.components under fault conditions. Certainversions can distinguish between resistiveand capacitive components of the leakagecurrent.Modern developments permit themeasurement of leakage-current evolution,so that prevention of a first fault can beachieved.Examples of equipment and devices**c manual fault-location (fig. G55).The generator may be fixed (example:XM200) or portable (example: XGRpermitting the checking of dead circuits) andthe receiver, together with the magnetic tong-type pick-up sensor, are portable.* On a 230/400 V 3-phase system.** The equipment and devices described to illustrate theprinciples of fault location, are manufactured by M.G.

P12 P50P100 ON/OFF

XGRXRM

XM200

MERLIN GERIN

XM100

fig. G55: non-automatic (manual) fault location.

c fixed automatic fault location (fig. G56)The monitoring relay XM200, together withthe fixed detectors XD301 (each suppliedfrom a toroidal CT embracing the conductorsof the circuit concerned) provide a system ofautomatic fault location on a live installation.Moreover, the level of insulation is indicatedfor each monitored circuit, and two levels arechecked: the first level warns of unusually lowinsulation resistance so that preventivemeasures may be taken, while the secondlevel indicates a fault condition and gives analarm.

XD301

XD312

1 to 12 circuits

toroidal CTs

XM200

MERLIN GERIN

XM100

XD301 XD301

fig. G56: fixed automatic fault location.

modern monitoring systems greatlyfacilitate first-fault location and repair.

G


6. implementation of the IT system (continued)


c automatic monitoring, logging, and faultlocation.The Vigilohm System also allows access to aprinter and/or a PC which provides a globalreview of the insulation level of an entireinstallation, and records the chronologicalevolution of the insulation level of eachcircuit.The central monitor XM300C, together withthe localization detectors XL308 and XL316,associated with toroidal CTs from severalcircuits, as shown below in figure G57,provide the means for this automaticexploitation.

XM300 C

MERLIN GERIN

XM100

XL308 XL316

MERLIN GERIN

XL08

897

MERLIN GERIN

XL16

678

fig. G57: automatic fault location and insulation-resistance data logging.

Implementation of permanent insulation-monitoring (PIM) devicesc connectionThe PIM device is normally connectedbetween the neutral (or articificial neutral)point of the power-supply transformer and itsearth electrode,c supplyPower supply to the PIM device should betaken from a highly reliable source. Inpractice, this is generally directly from theinstallation being monitored, throughovercurrent protective devices of suitableshort-circuit current rating,c impedance of the PIM deviceIn order to maintain the level of earth-faultwithin safe limits, the current passing througha PIM device during a short-circuit to earth isnormally limited to a value < 30 mA.Where the neutral point is earthed through animpedance, the total current passing throughthe PIM device and the impedance (inparallel with it) must be < 500 mA.This means that a touch voltage of less than50 V will occur in the installation as long asthe installation earth-electrode resistance isless than 100 ohms, and that fire risk ofelectrical origin is avoided,c level settingsCertain national standards recommend a firstsetting at 20% below the insulation level ofthe new installation. This value allows thedetection of a reduction of the insulationquality, necessitating preventive maintenancemeasures in a situation of incipient failure.The detection level for earth-fault alarm willbe set at a much lower level.

By way of an example, the two levels mightbe:v new installation insulation level: 100 kΩv leakage current without danger: 500 mA(fire risk at > 500 mA)v indication levels set by the consumer:- threshold for preventive maintenance:0.8 x 100 = 80 kΩ- threshold for short-circuit alarm: 300 mA.

Notes:v following a long period of shutdown, duringwhich the whole, or part of, the installationremains de-energized, humidity can reducethe general level of insulation resistance. Thissituation, which is mainly due to leakagecurrent over the damp surface of healthyinsulation, does not constitute a faultcondition, and will improve rapidly as thenormal temperature rise of current-carryingconductors reduces the surface humidity.v the PIM device (XM) can measure theresistive and the capacitive currentcomponents of the leakage current to earth,separately, thereby deriving the trueinsulation resistance from the total permanentcurrent leakage.

G


the case of a second faultA second earth fault on an IT system (unlessoccurring on the same conductor as the firstfault) constitutes a phase-phase or phase-to-neutral fault, and whether occurring on thesame circuit as the first fault, or on a differentcircuit, overcurrent protective devices (fusesor circuit breakers) would normally operate toeffect an automatic fault clearance.The settings of overcurrent tripping relaysand the ratings of fuses are the basicparameters that decide the maximumpractical length of circuit that can besatisfactorily protected, as discussed inSub-clause 5.2.Note: in normal circumstances, the faultcurrent path is through common PEconductors, bonding all exposed conductive

parts of an installation, and so the fault loopimpedance is sufficiently low to ensure anadequate level of fault current.Where circuit lengths are unavoidably long, andespecially if the appliances of a circuit areearthed separately (so that the fault currentpasses through two earth electrodes), reliabletripping on overcurrent may not be possible.In this case, an RCD is recommended oneach circuit of the installation.Where an IT system is resistance earthed,however, care must be taken to ensure thatthe RCD is not too sensitive, or a first faultmay cause an unwanted trip-out. Tripping ofresidual current devices which satisfy IECstandards may occur at values of 0.5 I∆n toI∆n, where I∆n is the nominal residual-currentsetting level.

at the remote end of a loop, when the level ofshort-circuit current at the near end of theloop is known. Complex impedances arecombined arithmetically in this method,c the conventional method, in which theminimum value of voltage at the origin of afaulty circuit is assumed to be 80% of thenominal circuit voltage, and tables are usedbased on this assumption, to give directreadings of circuit lengths.These methods are reliable only for the casesin which wiring and cables which make up thefault-current loop are in close proximity (toeach other) and are not separated by ferro-magnetic materials.

A reasonably accurate assessment of short-circuit current levels must be carried out atthe design stage of a project.A rigorous analysis is not necessary, sincecurrent magnitudes only are important for theprotective devices concerned (i.e. phaseangles need not be determined) so thatsimplified conservatively approximatemethods are normally used. Three practicalmethods are:c the method of impedances, based on thevectorial summation of all the (positive-phase-sequence) impedances around a fault-current loop,c the method of composition, which is anapproximate estimation of short-circuit current

three methods of calculating short-circuit current levels are commonlyemployed:c method of impedances, whichtakes account of complexrepresentation of impedances,c method of composition, is aconservatively approximate method,which combines impedancesarithmetically,c conventional method, is asimplified method based on anassumed minimum voltage duringfault, and the use of tables.

Method of impedancesThis method as described in Sub-clause 5.2,is identical for both the IT and TN systems ofearthing.

the software Ecodial 2 (Merlin Gerin)is based on the "method ofimpedances".

Method of compositionThis method as described in Sub-clause 5.2,is identical for both the IT and TN systems ofearthing.

Conventional methodThe principle is the same for an IT system asthat described in Sub-clause 5.2 for a TNsystem, viz: the calculation of maximumcircuit lengths which should not be exceededdownstream of a circuit breaker or fuses, toensure protection by overcurrent devices.It is clearly impossible to check circuit lengthsfor every feasible combination of twoconcurrent faults.All cases are covered, however, if theovercurrent trip setting is based on theassumption that a first fault occurs at theremote end of the circuit concerned, while thesecond fault occurs at the remote end of anidentical circuit, as already mentioned in Sub-clause 3.4. This may result, in general, in onetrip-out only occurring (on the circuit with thelower trip-setting level), thereby leaving thesystem in a first-fault situation, but with onefaulty circuit switched out of service.c for the case of a 3-phase 3-wire installationthe second fault can only cause a phase/phase short-circuit, so that the voltage to usein the formula for maximum circuit length iseUo.The maximum circuit length is given by:

Lm = 0.8 e Uo Sph metres 2 ρ (1+m) Ia

the maximum length of an IT earthedcircuit is:c for a 3-phase 3-wire scheme

L max = 0.8 Uo ex Sph 2 ρ Ia (1+m)c for a 3-phase 4-wire scheme

L max = 0.8 Uo S1 2 ρ Ia (1+m)

For the case of a 3-phase 4-wire installationthe lowest value of fault current will occur ifone of the faults is on a neutral conductor. Inthis case, Uo is the value to use forcomputing the maximum cable length, and,

Lm = 0.8 Uo S1 metres 2 ρ (1+m) Ia(i.e. 50% only of the length permitted for a TNscheme). Reminder: there is no length limitfor earth-fault protection on a TT scheme,since protection is provided by RCDs of highsensitivity.In the preceding formulae:Lmax = longest circuit in metresUo = phase-to-neutral voltage (230 V on a230/400 V system)ρ = resistivity at normal operatingtemperature= 22.5 x 10-3 ohms-mm2/m for copper= 36 x 10-3 ohms-mm2/m for aluminiumIa = overcurrent trip-setting level in ampsor Ia = current in amps required to clear thefuse in the specified timem = Sph/SPESPE = cross-sectional area of PE conductorin mm2

S1 = S neutral if the circuit includes a neutralconductor.

G




Id

PE CD

Id

AB

N

Id

PE

Id

N

fig. G58: calculation of Lmax. for an IT-earthed system, showing fault-current path for adouble-fault condition.

TablesThe following tables have been establishedaccording to the "conventional method"described above.The tables give maximum circuit lengths,beyond which the ohmic resistance of theconductors will limit the magnitude of theshort-circuit current to a level below thatrequired to trip the circuit breaker (or to blowthe fuse) protecting the circuit, with sufficientrapidity to ensure safety against indirectcontact. The tables take into account:c the type of protection: circuit breakers orfuses,c operating-current settings,c cross-sectional area of phase conductorsand protective conductors,c type of earthing scheme,c correction factor: table G59 indicates thecorrection factor to apply to the lengths givenin tables G43 to G46, when considering an ITsystem.

the following tables* give the lengthof circuit which must not beexceeded, in order that persons beprotected against indirect contacthazards by protective devices.

* The tables are those shown in Sub-clause 5.2 (tables G43to G46).However, the table of correction factors (table G59) whichtakes into account the ratio Sph/SPE, and of the type ofcircuit (3-ph 3-wire; 3-ph 4-wire; 1-ph 2-wire) as well asconductor material, is specific to the IT system, and differsfrom that for TN.

circuit conductor material m = S ph/S PE (or PEN)m = 1 m = 2 m = 3 m = 4

3 phases copper 0.86 0.57 0.43 0.34aluminium 0.54 0.36 0.27 0.21

3ph + N or 1ph + N copper 0.50 0.33 0.25 0.20aluminium 0.31 0.21 0.16 0.12

table G59: correction factors, for IT-earthed systems, to apply to the circuit lengths givenin tables G43 to G46.

ExampleA 3-phase 3-wire 230/400 V installation isIT-earthed.One of its circuits is protected by a circuitbreaker rated at 63 A, and consists of analuminium-cored cable with 50 mm2 phaseconductors. The 25 mm2 PE conductor is alsoaluminum. What is the maximum length ofcircuit, below which protection of personsagainst indirect-contact hazards is assured bythe instantaneous magnetic tripping relay ofthe circuit breaker?Table G44 indicates 617 metres, to whichmust be applied a correction factor of 0.36(m = 2 for aluminium cable).The maximum length is therefore 222 metres.

G


6.3 high-sensitivity RCDs



6.4 in areas of high fire-risk

RCD protection at the circuit breakercontrolling all supplies to the area at risk, ismandatory in many countries. The sensitivityof the RCD must be i 500 mA.

fire-risk area


G



6.5 when the fault-current-loop impedance is particularly high

When, during the design stage of theinstallation, it is found that the fault-currentloop impedance of a circuit will be inevitablyhigh, so that the overcurrent protectioncannot be relied upon to operate within theprescribed time, the following possibilitiesshould be considered:

Note: this is also the case when one (of two)earth faults occurs at the end of a longflexible lead, for example.

Suggestion 1:instal a circuit breaker which has aninstantaneous magnetic tripping element withan operation level which is lower than theusual setting, for example: 2In i Irm i 4 In.This affords protection on circuits which areabnormally long. It must be checked,however, that high transient currents such asthe starting currents of motors will not causenuisance trip-outs.

2In i Irm i 4In

PE or PEN

unusuallylong cable

fig. G62: a circuit breaker with low-setinstantaneous magnetic trip.

Suggestion 2:instal a RCD on the circuit of low sensitivity(several amps to a few tens of amps, since itmust not operate for a first fault).If the circuit is supplying socket outlets, it will,in any case, be protected by a high-sensitivityRCD (i 30 mA).

phasesneutralPE

TN-S

fig. G63: RCD protection.

Suggestion 3:increase the size of the PE conductors and/orthe phase conductors, to reduce the loopimpedance.

fig. G64: improved equipotential bonding.

Suggestion 4:add supplementary equipotential conductors.This will have a similar effect to that ofsuggestion 3, i.e. a reduction in the earth-fault-loop resistance, while at the same timeimproving the existing touch-voltageprotection measures. The effectiveness ofthis improvement may be checked by aresistance test between each exposedconductive part and the local main protectiveconductor.For TN-C installations, bonding as shown infigure G52 is not allowed, and Suggestion 3should be adopted.

G


7. residual current differential devices (RCDs)

7.1 description

principleThe essential features are showndiagrammatically in figure G65 below. Amagnetic core encompasses all the current-carrying conductors of an electric circuit andthe magnetic flux generated in the core willdepend at every instant on the arithmeticalsum of the currents; the currents passing inone direction being considered as positive,while those passing in the opposite directionwill be negative.In a normally healthy circuit (figure G65)i1 + i2 = 0 and there will be no flux in themagnetic core, and zero e.m.f. in its coil. Anearth-fault current id will pass through thecore to the fault, but will return to the source

via the earth, or via protective conductors in aTN-earthed system. The current balance inthe conductors passing through the magneticcore therefore no longer exists, and thedifference gives rise to a magnetic flux in thecore.The difference current is known as the"residual" current and the principle is referredto as the "differential current" principle.The resultant alternating flux in the coreinduces an e.m.f. in its coil, so that a currenti3 flows in the tripping-device operating coil. Ifthe residual current exceeds the valuerequired to operate the tripping device, thenthe associated circuit breaker will trip.

P

Ni1 i2

i3S

N

Ø

id

fig. G65: the principle of RCD operation.

7.2 application of RCDs

earth-leakage currents exist whichare not due to a fault, as well astransient overvoltages, either or bothof which can lead to unwantedtripping by RCDs.Certain techniques have beendeveloped to overcome theseoperational problems.

permanent earth leakage

currents

Every LV installation has a permanentleakage current to earth, which is mainly dueto imperfect insulation, and to the intrinsiccapacitance between live conductors andearth.The larger the installation the lower itsinsulation resistance and the greater itscapacitance with consequently increasedleakage current.On 3-phase systems the capacitive leakagecurrent to earth would be zero if theconductors of all three phases had equalcapacitance to earth, a condition that cannotbe realized in practical installations. Thecapacitive current to earth is sometimesincreased significantly by filtering capacitorsassociated with electronic equipment(automation, informatics and computer-basedsystems, etc.). In the absence of more-precise data, permanent leakage current in agiven installation can be estimated from thefollowing values, measured at 230 V 50 Hz,and abstracted from "Bulletin de l'UTE"April 1992.Fax terminal 0.5 to 1.0 mAIT* workstation 1 to 2 mAPrinter (IT*) < 1 mAIT* terminal 1 to 2 mAPhotocopier 0.5 to 1.5 mA* Information Technology.

transient leakage currents

The initial energization of the capacitancesmentioned above gives rise to high-frequencytransient currents of very short duration,similar to that shown in figure G66. Thesudden occurrence of a first-fault on an IT-earthed system also causes transient earth-leakage currents at high frequency, due to thesudden rise of the two healthy phases tophase/phase voltage above earth.

10 µs (f = 100 kHz)

t

100%90%

10%

ca.0.5 µs

60%

fig. G66: standardized 0.5 µs/100 kHzcurrent transient wave.

G


7. residual current differential devices (RCDs) (continued)

7.2 application of RCDs (continued)

influence of overvoltages

Electrical power networks are subjected toovervoltages of various origins; atmospheric,or due to abrupt changes of system operatingconditions (faults, fuse operation, switching,etc.). These sudden changes often causelarge transient voltages and currents insystem inductive and capacitive circuits,before a new stable state is reached.Records have established that, on LVsystems, overvoltages remain generallybelow 6 kV, and that they can be adequatelyrepresented by the conventional 1.2/50 µsimpulse wave (figure G67).These overvoltages give rise to transientcurrents represented by a current impulsewave of the conventional 8/20 µs form,having a peak value of several tens ofamperes (figure G68). The transient currentsflow to earth via the capacitances of theinstallation surge arresters or through aninsulation failure.

electromagnetic compatibility

The high-frequency (or unidirectionalimpulse) transient overvoltages and currentsmentioned above, together with otherelectromagnetic disturbance sources(contactor coils, relays, dry contacts),electrostatic discharges, and radiatedelectromagnetic waves (radio, ignitionsystems, etc.) are part of the increasinglyimportant field of EMC (electromagneticcompatibility). For further details, theTechnical publications nos. 120 and 149, byMerlin Gerin, may be consulted.It is essential that RCDs be immune topossible malfunction from the effects ofelectromagnetic-surge disturbances. Inpractice, the levels shown in table G70 arecomplied with in design and manufacturingspecifications*.* Merlin Gerin products.

implementation

c every RCD installed must have a minimumlevel of immunity to unwanted tripping inconformity with the requirements of tableG70. RCDs type "S" or time-delay settinglevels I or II (see figure G36) cover alltransient leakage currents, including those oflightning arresters (see installation layouts inChapter L, Sub-clause 1.3) of a duration lessthan 40 ms,c permanent leakage currents downstream ofa RCD must be studied, particularly in thecase of large installations and/or where filtercircuits are present, or again, in the case ofan IT-earthed installation. If the capacitancevalues are known, the equivalent leakage

fig. G67: standardized voltage-impulsewave 1.2/50 µs.

fig. G68: standardized current-impulsewave 8/20 µs.

fig. G69: standardized symbol used insome countries, to indicate proof againstincorrect operation due to transients.

current for the choice of the sensitivity of aRCD is: i mA* = 0.072 C at 50 Hzi mA = 0.086 C at 60 Hzwhere C = capacity (in n F) of one phase toearth.Since RCDs complying with IEC and manynational standards may operate within therange 0.5 I∆n - I∆n for a nominal rating of I∆n,the leakage current downstream of a RCDmust not exceed 0.5 I∆n.The limitation of permanent leakage currentto 0.25 I∆n, by sub-division of circuits, will, inpractice, eliminate the influence of allcorresponding current transients.For very particular cases, such as theextension, or partial renovation of extendedIT-earthed installations, the manufacturersmust be consulted.

i mA* = 230 V x 100 π x 103 C (n F) 109

i mA* = 0.072 C (n F) at 50 Hz

t

U

1.2 µs 50 µs

0.5 U

U max

t

0.5

0.9

0.1

8 µs

20 µs

I

table G70: electromagnetic compatibility withstand-level tests for RCDs.

disturbance type of test required withstand quantityovervoltage 1.2/50 µs impulse 6 kV peaktransient current 0.5 µs/100 kHz impulse 200 A peak*

8/20 µs impulse 200 A peak60 A peak for 10 mA RCDs5 kA peak for types "S"or time-delayed models (see note)

switching repetitive transient bursts IEC 801-4 4 kVstatic electricity electrostatic discharges IEC 801-2 8 kVradiated waves radiated electromagnetic fields IEC 801-3 3 V/m

* for RCDs having I∆n < 10 mA this test is not required (IEC 1008-1).Note: Time-delayed RCDs are normally installed near the service position of installations, where current surges of external originare the most severe. The 5 kA peak test reflects this high-performance duty requirement.

G


direct current components

Auxiliary d.c. supplies for control andindication of electrical and mechanicalequipment are common, and certainappliances include rectifiers (diodes, triacs,thyristors).In the event of an earth fault downstream of arectifier, the fault current can include a d.c.component.The risk depends on the level of insulation ofthe d.c. circuits in an appliance, and eachcase must be considered individually.Problems of this kind generally concernindustrial applications.The IEC classifies RCDs according to theirability to function correctly in the presence ofd.c. components in the residual current.

3 classes are distinguished:Class AC: operates due to a.c. current only.Class A: operates if residual current consistsof uni-directional pulses.Class B: operates on pure d.c.Note:For general use Class AC RCDs are normally installed.Class A are available for specific requirements as a specialvariation of Class AC devices.

recommendations concerning

the installation of RCDs with

separate toroidal current

transformers

The detector of residual current is a closedmagnetic circuit (usually circular) of very highmagnetic permeability, on which is wound acoil of wire, the ensemble constituting atoroidal (or ring-type) current transformer.Because of its high permeability, any smalldeviation from perfect symmetry of theconductors encompassed by the core, andthe proximity of ferrous material (steelenclosure, chassis members, etc.) can affectthe balance of magnetic forces sufficiently, attimes of large load currents (motor-startingcurrent, transformer energizing current surge,etc.) to cause unwanted tripping of the RCD.Unless particular measures are taken, theratio of operating current I∆n to maximumphase current Iph (max.) is generally lessthan 1/1,000.This limit can be increased substantially(i.e. the response can be desensitized) byadopting the measures shown in fig. G71,and summarized in table G72.

Centralize the cables in the ring core

Use an oversized magnetic ring core

Insert a tubular magnetic screen.

L

L = twice the diameter of the magnetic ring core

fig. G71: means of reducing the ratioI∆n/Iph (max.).

measures diameter (mm) sensitivitydiminutionfactor

careful centralizing of cables through the ring core 3oversizing of the ring core ø 50 > ø 100 2

ø 80 > ø 200 2ø 120 > ø 200 6ø 50 4ø 80 3ø 120 3ø 200 2

These measures can be combined. By carefully centralizing the cables in a ring core of 200 mmdiameter, where a 50 mm core would be large enough, and using a sleeve, the ratio 1/1,000could become 1/30,000.

use of a steel or soft-iron shielding sleevec of wall thickness 0.5 mmc of length 2 x inside diameter of ring corec completely surrounding the conductors and overlappingthe circular core equally at both ends

table G72: means of reducing the ratio I ∆n/Iph (max.).

G


7. residual current differential devices (RCDs) (continued)

7.3 choice of characteristics of a residual-current circuit breaker (RCCB - IEC 1008)

rated currentThe rated current of a RCCB is chosenaccording to the maximum sustained loadcurrent it will carry, estimated in accordancewith the methods described in Chapter BSub-clause 4.3.c if the RCCB is connected in series with,and downstream of a circuit breaker, therated current of both items will be the same,i.e. In u In1* (fig. G73 (a)),c if the RCCB is located upstream of a groupof circuits, protected by circuit breakers, asshown in fig. G73 (b), then the RCCB ratedcurrent will be given byIn u ku x ks (In1 + In2 + In3 + In4).* Some national standards include a thermal withstand testat a current greater than In in order to ensure correctcoordination of protection.

In1

In

(a)

In1

In

In2 In3 In4

(b)

fig. G73: residual current circuit breakers(RCCBs).

electrodynamic withstand

requirements

Protection against short-circuits must beprovided by an upstream SCPD (Short-CircuitProtective Device) but it is considered thatwhere the RCCB is located in the samedistribution box (complying with theappropriate standards) as the downstreamcircuit breakers (or fuses), the short-circuitprotection afforded by these (outgoing-circuit)SCPDs is an adequate alternative.Coordination between the RCCB and theSCPDs is necessary, and manufacturersgenerally provide tables associating RCCBsand circuit breakers or fuses (see table G74).

Coordination of circuit breakers and RCCBs- max. short-circuit current in kA (r.m.s.)upstream circuit breaker type C60a C60N C60H C60L NC100H NC100Ldownstream 2 p 25 A 10 16 20 45 45RCCB 40 A 10 16 20 40 45

63 A 16 20 30 5 4580 A 5

4 p 25 A 5 8 10 25 2240 A 5 8 10 25 2263 A 8 10 15 5 22

Coordination of fuses and RCCBs- max. short-circuit (not applicable to aM fuses)upstream fuses gl(not applicable to aM fuses) 16 A 25 A 32 A 40 A 50 A 63 A 80 A 100 Adownstream 2 p 25 A 100 100 100RCCB 40 A 100 100 80 10 (1)

63 A 80 50 30 20 10 (1)

80 A 30 204 p 25 A 100 100 100 10 (1)

40 A 100 100 80 10 (1)

63 A 80 50 30 20 10 (1)

80 A 30 20 10 (1)

table G74: typical manufacturers coordination table for RCCBs, circuit breakers,and fuses.(1) A 100 A fuse with several RCCBs downstream: the thermal withstand of the RCCBs is not certain.

the protection of circuits - the switchgear

H

The protection of circuitsthe switchgear

1. General1.1 methodology and definitions H1-11.2 overcurrent protection principles H1-31.3 practical values for a protection scheme H1-41.4 location of protective devices H1-51.5 worked example of cable calculations H1-6

2. Practical method for determining the smallestallowable cross-sectional-area of circuit conductors2.1 general H1-102.2 determination of conductor size for unburied circuit H1-122.3 determination of conductor size for buried circuit H1-14

3. Determination of voltage drop3.1 maximum voltage-drop limit H1-173.2 calculation of voltage drops in steady load conditions H1-18

4. Short-circuit current4.1 short-circuit current at the secondary terminals H1-20of a HV/LV distribution transformer4.2 3-phase short-circuit current (Isc) at any point within an installation H1-214.3 Isc at the receiving end of a feeder in terms of the Isc at its sending end H1-234.4 short-circuit current supplied by an alternator or an inverter H1-25

5. Particular cases of short-circuit current5.1 calculation of minimum levels of Isc H1-265.2 verification of the withstand capabilities H1-31of cables under short-circuit conditions

6. Protective earthing conductors (PE)6.1 connection and choice H1-326.2 conductor dimensioning H1-336.3 protective conductor between the HV/LV transformer H1-35and the main general distribution board (MGDB)6.4 equipotential conductor H1-35

7. The neutral conductor7.1 dimensioning the neutral conductor H1-367.2 protection of the neutral conductor H1-37

1. The basic functions of LV switchgear1.1 electrical protection H2-11.2 isolation H2-11.3 switchgear control H2-2

2. The switchgear2.1 elementary switching devices H2-62.2 combined switchgear elements H2-9

3. Choice of switchgear3.1 tabulated functional capabilities H2-113.2 switchgear selection H2-11

4. Circuit breakers4.1 standards and descriptions H2-124.2 fundamental characteristics of a circuit breaker H2-154.3 other characteristics of a circuit breaker H2-184.4 selection of a circuit breaker H2-204.5 co-ordination between circuit breakers H2-274.6 discrimination HV/LV in a consumer's substation H2-32

H1

H2

the protection of circuits - the switchgear - H1-1

H1

1. general

1.1 methodology and definitions

component parts of an electric circuitand its protection are determinedsuch, that all normal and abnormaloperating constraints are satisfied.

methodologyFollowing a preliminary analysis of the powerrequirements of the installation, as decribedin Chapter B Clause 4, a study of cabling*and its electrical protection is undertaken,starting at the origin of the installation,through the intermediate stages to the finalcircuits.The cabling and its protection at each levelmust satisfy several conditions at the sametime, in order to ensure a safe and reliableinstallation, e.g. it must:c carry the permanent full load current, andnormal short-time overcurrents,c not cause voltage drops likely to result in aninferior performance of certain loads, forexample: an excessively long accelerationperiod when starting a motor, etc.Moreover, the protective devices (circuitbreakers or fuses) must:c protect the cabling and busbars for alllevels of overcurrent, up to and includingshort-circuit currents,

c ensure protection of persons againstindirect contact hazards, particularly inTN- and IT- earthed systems, where thelength of circuits may limit the magnitudeof short-circuit currents, thereby delayingautomatic disconnection (it may beremembered that TT- earthed installations areobligatorily protected at the origin by a RCD,generally rated at 500 mA).The cross-sectional areas of conductors aredetermined by the general method describedin Sub-clause 1.2 of this Chapter. Apart fromthis method some national standards mayprescribe a minimum cross-sectional area tobe observed for reasons of mechanicalendurance. Particular loads (as noted inChapter J) require that the cable supplyingthem be oversized, and that the protection ofthe circuit be likewise modified.

* the term "cabling" in this chapter, covers all insulatedconductors, including multi-core and single-core cables andinsulated wires drawn into conduits, etc.

kVA to be supplied short-circuit MVA at the origin of the circuit

rated current of protectivedevice (C.B. or fuses)

short-circuit current-breaking rating of C.B. or fuses

choice of C.B. or fuses

verification of thermal withstand requirements

verification of the maximum voltage drop

IT or TN scheme

TT scheme

verification of the maximum length of the circuit

confirmation of the cross-sectional area of the cabling, and the choice of its electrical protection

determination of the cross-sectional areaof the conductors

choice of protective device

upstream or downstream network

cross-sectional area of conductors of the circuit

maximum load current

short-circuitcurrent

conditions of installation

IB

In

Isc

Iscb

table H1-1: logigram for the selection of cable size and protective-device ratingfor a given circuit.

H1-2 - the protection of circuits - the switchgear

H1


1.1 methodology and definitions (continued)

definitionsMaximum load current: I Bc at the final circuits level, this currentcorresponds to the rated kVA of the load. Inthe case of motor-starting, or other loadswhich take an initially-high current,particularly where frequent starting isconcerned (e.g. lift motors, resistance-typespot welding, and so on) the cumulativethermal effects of the overcurrents must betaken into account. Both cables and thermal-type relays are affected;c at all upstream circuit levels this currentcorresponds to the kVA to be supplied, whichtakes account of the factors of simultaneity(diversity) and utilization, ks and kurespectively, as shown in figure H1-2.Maximum permissible current: I ZThis is the maximum value of current that thecabling for the circuit can carry indefinitely,without reducing its normal life expectancy.The current depends, for a given cross-sectional area of conductors, on severalparameters:c constitution of the cable and cable-way(Cu or Alu conductors; PVC or EPR etc.insulation; number of active conductors);c ambient temperature;c method of installation;c influence of neighbouring circuits.

maindistributionboard

sub-distributionboard

80 A 60 A 100 A IB = 50 A

Mnormal loadmotor current50 A

combined factorsof simultaneity(or diversity)and utilizationks x ku = 0.69

IB = 290 x 0.69 = 200 A

fig. H1-2: calculation of maximum loadcurrent I B.

overcurrentsAn overcurrent occurs each time the value ofcurrent exceeds the maximum load current IBfor the load concerned.This current must be cut off with a rapiditythat depends upon its magnitude, ifpermanent damage to the cabling (andappliance if the overcurrent is due to adefective load component) is to be avoided.Overcurrents of relatively short duration canhowever, occur in normal operation; twotypes of overcurrent are distinguished:

OverloadsThese overcurrents can occur in healthyelectric circuits, for example, due to a numberof small short-duration loads whichoccasionally occur co-incidentally; motor-starting loads, and so on.If either of these conditions persists howeverbeyond a given period (depending onprotective-relay settings or fuse ratings) thecircuit will be automatically cut off.

Short-circuit currentsThese currents result from the failure ofinsulation between live conductors or/andbetween live conductors and earth (onsystems having low-impedance-earthedneutrals) in any combination, viz:c 3 phases short-circuited (and to neutraland/or earth, or not);c 2 phases short-circuited (and to neutraland/or earth, or not);c 1 phase short-circuited to neutral (and/or toearth).


H1

1.2 overcurrent protection principles

A protective device is provided at the origin ofthe circuit concerned.c acting to cut-off the current in a time shorterthan that given by the I2t characteristic of thecircuit cabling;c but allowing the maximum load current IB toflow indefinitely.The characteristics of insulated conductorswhen carrying short-circuit currents can, forperiods up to 5 seconds following short-circuitinitiation, be determined approximately by theformula:Is2 x t = k2 x S2 which shows that theallowable heat generated is proportional tothe cross-sectional-area of the condutorsquared.Where:t: duration of short-circuit current (seconds);S: c.s.a. of insulated conductor (mm2);Is: short-circuit current (A r.m.s.);k: insulated conductor constant (valuesof k2 are given in table H1-54).For a given insulated conductor, themaximum permissible current variesaccording to the environment. For instance,for a high ambient temperature (θa1 > θa2),IZ1 is less than IZ2 (fig. H1-5).θ means "temperature".

Note:Isc means 3-phase short-circuit current.IscB means rated 3-ph. short-circuit breakingcurrent of the circuit breaker.Ir (or Irth)* means regulated "nominal" currentlevel; e.g. a 50 A nominal circuit breaker canbe regulated to have a protective range, i.e.a conventional overcurrent tripping level(see figure H1-6) similar to that of a 30 Acircuit breaker.* both designations are commonly used in differentstandards.

t

IIB Ir Iz ISCB PdC

maximumloadcurrent

I2t cablecharacteristic

circuit-breakertripping curve

temporaryoverload

fig. H1-3: circuit protection by circuitbreaker.

t

I

I2t cablecharacteristic

fusecurve

IB Ir cIz Iz

temporaryoverload

fig. H1-4: circuit protection by fuses.

t

I

θa1 > θa2

Iz1 < Iz2

I2t = k2S25 s

1 2

fig. H1-5: I 2t characteristic of an insulatedconductor at two different ambienttemperatures.


H1

3-ph

shor

t-circ

uit

fault-c

urren

t brea

king r

ating

loads circuit cabling

max

imum

per

miss

ible

curre

nt I z

1.45

x I z

maxim

um load current IB

nom

inal c

urre

nt I n

or

its re

gulat

ed cu

rrent

I r

protective device

ISCBzone a

conv

entio

nal o

verc

urre

nt

trip

curre

nt I 2

zone b zone c

IB Iz

In I2

1.45 Iz ISC


1.3 practical values for a protection scheme

The following methods are based on ruleslaid down in the IEC standards, and are

representative of the practices in manycountries.

fig. H1-6: current levels for determining circuit breaker or fuse characteristics.

general rulesA protective device (circuit breaker or fuse)functions correctly if:c its nominal current or its setting current In isgreater than the maximum load current IB butless than the maximum permissible current IZfor the circuit, i.e. IB i In i IZ corresponding tozone "a" in figure H1-6;c its tripping current I2 "conventional" settingis less than 1.45 IZ which corresponds tozone "b" in figure H1-6.

IB i In i Iz zone aI2 i 1,45 Iz zone bISCB u ISC zone c

The "conventional" setting tripping time maybe 1 hour or 2 hours according to localstandards and the actual value selected forI2.For fuses, I2 is the current (denoted If) whichwill operate the fuse in the conventional time;c its 3-phase short-circuit fault-currentbreaking rating is greater than the 3-phaseshort-circuit current existing at its point ofinstallation.This corresponds to zone "c" in figure H1-6.

applicationsProtection by circuit breakerBy virtue of its high level of precision thecurrent I2 is always less than 1.45 In (or1.45 Ir) so that the condition, that I2 i 1.45 IZ(as noted in the "general rules" above) willalways be respected.

Particular case:if the circuit breaker itself does not protectagainst overloads, it is necessary to ensurethat, at a time of lowest value of short-circuitcurrent, the overcurrent device protecting thecircuit will operate correctly. This particularcase is examined in Sub-clause 5.1.

criteria for a circuit breaker:IB i In (or Ir) i Izand,rated short-circuit breaking currentISCB u ISC the 3-ph. short-circuitcurrent level at the point of CBinstallation.

Protection by fusesThe condition I2 i 1.45 IZ must also be takeninto account, where I2 is the fusing (melting-level) current, equal to k2 x In (k2 ranges from1.6 to 1.9) according to the particular fuseconcerned.A further factor k3 has been introduced (in thenational standards from which these noteshave been abstracted) such that I2 i 1.45 IZwill be valid if In i IZ/k3.

For fuses type gl:In i 10 A k3 = 1.3110 A < In i 25 A k3 = 1.21In > 25 A k3 = 1.10Moreover, the short-circuit current breakingcapacity of the fuse ISCF must exceed thelevel of 3-phase short-circuit current at thepoint of installation of the fuse(s).

criteria for fuses:

IB i In i IZ k3and,the rated short-circuit currentbreaking capacity of the fuseISCF u ISC the 3-ph. short-circuitcurrent level at the point of fuseinstallation.

Association of different protective devicesThe use of protective devices which havefault-current ratings lower than the fault levelexisting at their point of installation arepermitted by IEC and many nationalstandards in the following conditions:c there exists upstream, another protectivedevice which has the necessary short-circuitrating, andc the amount of energy allowed to passthrough the upstream device is less than thatwhich the downstream device and all

associated cabling and appliances canwithstand without damage.In pratice this arrangement is generallyexploited in:c the association of circuit breakers/fuses;c the technique known as "cascading" inwhich the strong current-limiting performanceof certain circuit breakers effectively reducesthe severity of downstream short-circuits.Possible combinations which have beentested in laboratories are indicated in certainmanufacturers catalogues.


H1

1.4 location of protective devices

a protective device is, in general,required at the origin of each circuit.

general ruleA protective device is necessary at the originof each circuit where a reduction ofpermissible maximum current level occurs.

P

P2 P3 P4

50 mm2 10 mm2 25 mm2

P2

P3

(1) (2)

short-circuitprotectivedevice

P1

sc

sB

overloadprotectivedevice3

B

B

< 3 m

A

case (3)case case

possible alternative locations in

certain circumstances

The protective device may be placed partway along the circuit:c if AB is not in proximity to combustiblematerial, andc if no socket-outlets or branch connectionsare taken from AB.Three cases may be useful in practice.Consider case (1) in the diagramc AB i 3 metres, andc AB has been installed to reduce to apractical minimum the risk of a short-circuit(wires in heavy steel conduit for example).

Consider case (2)c the upstream device P1 protects the lengthAB against short-circuits in accordance withSub-clause H1-5.1.

Consider case (3)c the overload device (S) is located adjacentto the load. This arrangement is convenientfor motor circuits. The device (S) constitutesthe control (start/stop) and overloadprotection of the motor while (SC) is: either acircuit breaker (designed for motor protection)or fuses type aM,c the short-circuit protection (SC) located atthe origin of the circuit conforms with theprinciples of Sub-clause H1-5.1.

P1: C60 calibre 15 A

S2:1,5 mm2

2,5 mm2

circuits with no protection

Eitherc the protective device P1 is calibrated toprotect the cable S2 against overloads andshort-circuits;

Orc where the breaking of a circuit constitutes arisk, e.g.v excitation circuits of rotating machines,v circuits of large lifting electromagnets,v the secondary circuits of currenttransformers.No circuit interruption can be tolerated, andthe protection of the cabling is of secondaryimportance.

table H1-7: general rules and exceptions concerning the location of protective devices.

1.5 cables in parallel

Conductors of the same cross-sectional-area,the same length, and of the same material,can be connected in parallel.The maximum permissible current is the sumof the individual-core maximum currents,taking into account the mutual heatingeffects, method of installation, etc.Protection against overload and short-circuitsis identical to that for a single-cable circuit.

The following precautions should be taken toavoid the risk of short-circuits on theparalleled cables:c additional protection against mechanicaldamage and against humidity, by theintroduction of supplementary protection;c the cable route should be chosen so as toavoid close proximity to combustiblematerials.


H1


1.6 worked example of cable calculations

installation scheme

The installation is supplied through a1,000 kVA transformer. The process requiresa high degree of supply continuity and this isprovided by the installation of a 500 kVA400 V standby generator, and by the adoptionof a 3-phase 3-wire IT-system at the maingeneral distribution board from which theprocessing plant is supplied. The remainderof the installation is isolated by a 315 kVA400/400V transformer: the isolated network isa TT-earthed 3-phase 4-wire system.

Following the one-line diagram of the systemshown in figure H1-8 below, a reproduction ofthe results of a computer study for the circuitC1 and its circuit breaker Q1, and C2 withassociated circuit breaker Q2 are presented.These studies were carried out withECODIAL 2.2 software (a Merlin Gerinproduct).This is followed by the same calculationscarried out by the methods described in thisguide.

Q6Q5

Q8 Q9 Q10

Q4Q3

TR11000 kVA5%

400 V26. 44 kA

3x (3 x 240)

C18m.18%

Q1M16 N1STR 381600 A

B1

Q2C801NSTR35SE800 A

C215m.7%

3x (1 x 240)

C3

T1315 kVA

400 V

Q7NS630NSTR35SE630 A

B2

G1500 kVA721 A

C4

I1 I2

Q11NS250NTMD250 A

Q12NS160NTMD160 A

Q13NS100NTMD80 A

fig. H1-8: one-line diagram of the installation.


H1

calculations using software Ecodial 2.2

General network characteristicsearthing system ITneutral distributed Nvoltage (V) 400frequency (Hz) 50

Transformer TR 1 input data outputnumber of transformers 1upstream fault level (MVA) 500rating (kVA) 1000short-circuit impedance voltage (%) 5remarksnominal current (A) 1374resistance of transformer (mΩ) 2.13reactance of transformer (mΩ) 8.55running total of impedance RT (mΩ) 2.18running total of impedance XT (mΩ) 8.93-phase short-circuit current (kA) 26.44short-circuit power factor .23

Cable C 1 input data outputmaximum load current (A) 1374type of insulation PRCconductor material Cuambient temperature (°C) 30single-core or multi-core cable UNIinstallation method 13number of circuits in close proximity(table H1-14) 1other coefficient 1number of phases 3selected cross-sectional area (mm2) 3 x 240protective conductor 1 x 240neutral conductorlength (m) 8voltage drop ∆U (%) .18running total of impedance RT (mΩ) 2.43running total of impedance XT (mΩ) 9.11voltage drop ∆U total (%) .183-phase short-circuit current (kA) 25.71-phase-to-earth fault current (A) 20334resistance of protective conductor RPE (mΩ) .75touch voltage (V) 15

Circuit breaker Q 1 input data outputvoltage (V) 4003-ph short-circuit current upstreamof the circuit breaker (kA) 25.7maximum load current (A) 1374ambient temperature (°C) 40number of poles 3circuit breaker M 16type N 1tripping unit type STR 38rated current (A) 1600

Busbars B 1maximum load current (A) 1374number of phases 3number of bars per phase 1width (mm) 125thickness (mm) 5length (m) 3remarksimpedance of busbars R (mΩ) .1impedance of busbars X (mΩ) .45voltage drop ∆U(%) .16running total of impedance RT (mΩ) 2.53running total of impedance XT (mΩ) 9.55voltage drop ∆U total (%) .343-ph short-circuit current (kA) 24.53


H1


1.6 worked example of cable calculations (continued)

Circuit breaker Q 2 input data outputvoltage (V) 4003-ph short-circuit current upstreamof the circuit breaker (kA) 24.53maximum load current (A) 433ambient temperature (°C) 40number of poles 3circuit breaker NS630type Ntripping unit type STR23SErated current (A) 6303-phase fault current (A) 13221protection against indirect contact assuredupstream circuit breaker M16 N1 STR38absolute discrimination

Cable C 2 input data outputmaximum load current (A) 433type of insulation PRCconductor material Cuambient temperature (°C) 30single-core or multi-core cable UNIinstallation method 13number of circuits in close proximity(table H1-14) 1other coefficient 1number of phases 3selected cross-sectional area (mm2) 1 x 240protective conductor 1 x 70neutral conductorlength (m) 15voltage drop ∆U (%) .33running total of impedance RT (mΩ) 3.93running total of impedance XT (mΩ) 10.75voltage drop ∆U total (%) .673-phase short-circuit current (kA) 21.181-phase-to-earth fault current (A) 13221resistance of protective conductor RPE (mΩ) 5.57touch voltage (V) 73

table H1-9: calculations carried out with ECODIAL software (M.G).

the same calculations using

the methods recommended in

this guide

Dimensioning circuit C 1The HV/LV 1,000 kVA transformer has a ratedno-load voltage of 420 V. Circuit C 1 must besuitable for a current of 1,000In =

ex 0.42 = In = 1,374 A per phase

Three single-core XLPE-insulated coppercables in parallel will be used for each phase;these cables will be laid on cable trayscorresponding with reference F (see tables inClause H1 2.2). The "K" correction factors areas follows:K 1 = 1K 2 = 0.82 ( 3 three-phase groups in a singlelayer)K 3 = 1 (temperature 30 °C).If the circuit breaker is a withdrawable orunpluggable* type, which can be regulated,one might choose:Iz = 1,374 A applying H1.2.1

I’z = Iz = 1,676 A. K 1 x K 2 x K 3Each conductor will therefore carry 558 A.Table H1-17 indicates that the c.s.a. is240 mm2.* Withdrawable, CBs are generally mounted in drawers formaintenance purposes. Plug-in type CBs are generallymoulded-case units, which may be completely removedfrom the fixed-base sockets.

The resistances and the inductive reactancesfor the three conductors in parallel are, for alength of 8 metres (see H1-4.2):

R = 22.5 x 8 = 0.25 mΩ per phase 240 x 3

X = 0.12 x 8 = 0.32 mΩ per phase 3(0.12 mΩ/metre was advised by the cablemaker).Dimensioning circuit C 2Circuit C 2 supplies a 315 kVA 3-phase400/400 V isolating transformer

Ib = 315 0.42 x e

= 433 A.

A multi-core XLPE cable laid on a cable tray(together with two other cables) in an ambientair temperature of 30 °C is proposed.The circuit breaker is regulated to 433 A.Iz = 433 AThe method of installation is characterized bythe reference letter E, and the "K" correctingfactors are:K 1 = 1K 2 = 0.82K 3 = 1.

I’z = 433 = 528 A so that 1 x 0.82 x 1a c.s.a. of 240 mm2 is appropriate.The resistance and inductive reactance arerespectively:

R = 22.5 x 15 = 1.4 mΩ per phase 240X = 0.08 x 15 = 1.2 mΩ per phase.


H1

Sub-clause H1-4.2 shows the formula forcalculating the short-circuit current Isc at agiven point in the system.If the rated no-load voltage of the transformeris 420 V:

= 26.5 kA at Q 1.The inductive reactance of busbars B1 isestimated to be 0.15 x 5 = 0.75 mΩ - itsresistance being negligibly small.The Isc at the location of Q 2 is computed asfor Q 1, and found to be 21 kA.In order to make the final choice, featuressuch as selectivity, isolating capability,withdrawal or unplugging facility and generalease of maintenance, and so on, must beconsidered, with the aid of manufacturerscatalogues.

The protective conductorThermal requirements.Tables H1.60 and H1.61 show that, whenusing the adiabatic method (IEC 724 (1984)Clause 2) the c.s.a. for the protective earth(PE) conductor for circuit C1 will be:

u 26500 x √0.1 = 47.6 mm2

176a single 240 mm2 conductor dimensioned forother reasons mentioned later is thereforelargely sufficient, provided that it also satisfiesthe requirements for indirect-contactprotection (i.e. that its impedance issufficiently low).For the circuit C2, the c.s.a. of its PEconductor should be:

u 21,000 x √0.1 = 37.7 mm2

176In this case a 70 mm2 conductor may beadequate if the indirect-contact protectionconditions are also satisfied.Protection against indirect-contacthazardsReminder: the LV neutral point of anIT-scheme transformer is isolated from earth,or is earthed through a high resistance(1-2 kΩ) so that an indirect-contact hazardcan only exist if two earth faults occurconcurrently, each on a different phase (or onone phase and a neutral conductor).Overcurrent protective devices must then berelied upon to cut-off the faulty circuits, exceptin particular circumstances i.e. where theresistance of P.E. conductors is too high, asnoted in Chapter G Sub-clauses 6.3 to6.5. RCDs are often employed in such cases.

=Isc 420

2.542 + 8.7723

Calculation of short-circuit currents forthe selection of circuit breakers Q 1 andQ 2

*all values are to a 420 voltage base ecircuits R* X* Z* Isc*components parts m Ω mΩ mΩ kA500 kVA at theHV source network 0.050 0.35HV/LV transformer 2.24 8.10cable C 1 0.25 0.32sub-total for Q1 2.54 8.77 9.13 26.5busbars B1 - 0.75 9.85 24.6cable C 2 1.40 1.2sub-total for Q2 3.94 10.72 11.42 21.2table H1-10: example of short-circuitcurrent evaluation.

Circuit C 1 will be of class 2 insulation i.e.double insulation and no earthed exposedconductive parts. The only indirect-contactrequirement for this circuit, therefore, is at thetransformer tank. The 240 mm2 P.E.conductor mentioned above, generallyconnects the tank of the HV/LV transformer tothe earth electrode for the installation at acommon earthing busbar in the main generaldistribution board. This means that if one (ofthe two) concurrent LV phase-to-earth faultsshould occur in the transformer, an indirectcontact danger will exist at the transformertank.In such a case, the HV overcurrent protectionfor the transformer is unlikely to operate, butthe protection on the second faulty LV circuitmust do so infallibly to ensure protectionagainst the indirect contact danger, asdescribed.Since a HV fault to earth at the transformer isalso always possible, and very often HVlightning arresters on the transformer areconnected to earth through the P.E.conductor in question, a conductor of largec.s.a. is invariably selected for this sectionof the installation. Dimensioningconsiderations for this conductor are givenin Sub-clause 6.3.For circuit C2, tables G.43 and G.59, or theformula given in Sub-clause G.6.2 may beused for a 3-phase 3-wire circuit.The maximum permitted length of the circuitis given by:

Lmax = 0.8 x 230 x 240 x ex 103

2 x 22.5 x (1.25 + 240/70) x 630 x 11.5

= 76,487 = 50 metres. 1,530The factor 1.25 in the denominator is a 25%increase in resistance for a 240 mm2

conductor, in accordance with Chapter GSub-clause 5.2.(The value in the denominator630 x 11.5 = Im i.e. the current level at whichthe instantaneous short-circuit magnetic tripof the 630 A circuit breaker operates). Thisvalue is equal to 10 In + 15 % (the highestpositive manufacturing tolerance for thetripping device). For further details ofmagnetic tripping devices, please refer toChapter H2 Sub-clause 4.2.The length of 15 metres is therefore fullyprotected by "instantaneous" overcurrentdevices.

Voltage dropFrom table H1.29 it can be seen that:for C1 (3 x 240 mm2 per phase)

∆U = 0.21 V/A/km x 1,374 A x 0.008 km 3= 0.77 V

∆U % = 100 x 0.77 = 0.19 %; 400for C2∆U = 0.21 V/A/km x 433 A x 0.015 km= 1.36 V

∆U % = 100 x 1.36 = 0.34 %; 400At the circuit terminals of the LV/LVtransformer the percentage volt-drop∆U % = 0.53 %.


H1

2. practical method for determining the smallest allowablecross-sectional-area of circuit conductors

2.1 general

installation conditions for the conductors

maximum load current IB

circuit breaker

choice of maximum permissible current IZ for the circuit, corresponding to a conductor size that the protective device is capable of protecting

fuse

and In 25 A*

rated current In of the protective device must be equal to or greater than the maximum load current IB

Determination of the size (c.s.a.) of the conductors of the circuit capable of carrying IZ1 or IZ2, by use of an equivalent current I'Z, which takes into account the influences of factor K (I'Z = IZ/K), ofthe letter code, and of the insulating sheath of the conductors (refer to tables H1-17 or H1-24)

verification of other conditions that may be required-see figure H1.1

I = IZ

I Z1 I Z2

S2S1

I n

I B

n*

I 'Z I 'Z

* or slightly greater

determination of K factors and of the appropriate letter code

IZ = 1.31 In if In 10 A*IZ = 1.21 In if In 10 A*

IZ = 1.10 In if In 25 A*

table H1-11: logigram for the determination of minimum conductor size for a circuit.

The first step is to determine the size of thephase conductors. The dimensioning of theneutral and protective conductors isexplained in H1-6 and H1-7.In this clause the following cases areconsidered:c unburied conductors,c buried conductors.The tables in this clause permit thedetermination of the size of phase conductorsfor a circuit of given current magnitude.The procedure is as follows:c determine an appropriate code-letter

reference which takes into account:v the type of circuit (single-phase; three-phase, etc.) andv the kind of installation: and thenc determine the factor K of the circuitconsidered, which covers the followinginfluences:v installation method,v circuit grouping,v ambient temperature.

2.2 determination of conductor size for unburied circuits

the size of a phase conductor isgiven in tables which relate:c the code letter symbolizing themethod of installation, andc the factor of influence K.These tables distinguish unburiedcircuits from buried circuits.

determination of the code-letter

reference

The letter of reference (B to F) depends onthe type of conductor used and its method ofinstallation. The possible methods of

installation are numerous, but the mostcommon of them have been groupedaccording to four classes of similarenvironmental conditions, as shown below intable H1-12.

types of conductor method of installation letter codesingle-core wires and c under decorative moulding with ormulti-core cables without a removable cover, surface

or flush-mounting, or under plasterc in underfloor cavity or behind Bfalse ceilingc in a trench, moulding or wainscotingc surface-mounted in contact withwall or ceilingc on non-perforated cable trays C

multi-core cables c cable ladders, perforated trays, Eor on supporting bracketsc surface-mounted clear of the surface(e.g. on cleats)c catenary cables

single-core cables F

table H1-12: code-letter reference, depending on type of conductor and method ofinstallation.


H1

for circuits which are not buried,factor k characteristizes theconditions of installation, and is givenby: K = K1 x K2 x K3the three component factorsdepending on different featuresof the installation.

determination of the factor KThe factor k summarizes the several featureswhich characterize the conditions ofinstallation.It is obtained by multiplying three correctionfactors K1, K2 and K3.The values of these factors are given intables H1.13 to H1.15 below.

correction factor K1

Factor K1 is a measure of the influence of themethod of installation.

code letter installation details example K1B - cables installed directly in 0.70

thermal-insulation materials

- conduits installed in thermal- 0.77insulation materials

- multi-core cables 0.90

- construction cavities and closed 0.95cables trenches

C - surface mounted on ceiling 0.95

B, C, E, F - other cases 1

factor K1 is a measure of theinfluence of the method ofinstallation.

table H1-13: factor K1 according to method of circuit installation (for further examplesrefer to IEC 364-5-52 table 52H).

Correction factor K2Factor K2 is a measure of the mutualinfluence of two circuits side-by-side in closeproximity.

Two circuits are considered to be in closeproximity when L, the distance betweentwo cables, is less than double the diameterof the larger of the two cables.

factor K2 is a measure of the mutualinfluence of two circuits side-by-sidein close proximity.

code location of correction factor K2letter cables in close number of circuits or multicore cables

proximity 1 2 3 4 5 6 7 8 9 12 16 20B,C embedded 1.00 0.80 0.70 0.65 0.60 0.57 0.54 0.52 0.50 0.45 0.41 0.38

or buriedin the walls

C single layer 1.00 0.85 0.79 0.75 0.73 0.72 0.72 0.71 0.70 0.70on walls orfloors, or onunperforatedcables trayssingle layer 0.95 0.81 0.72 0.68 0.66 0.64 0.63 0.62 0.61 0.61on ceiling

E,F single layer 1.00 0.88 0.82 0.77 0.75 0.73 0.73 0.72 0.72 0.72on horizontalperforatedtrays, or onvertical trayssingle layer 1.00 0.87 0.82 0.80 0.80 0.79 0.79 0.78 0.78 0.78on cable ladders,brackets, etc

table H1-14: correction factor K2 for a group of conductors in a single layer


H1

2. practical method for determining the smallest allowablecross-sectional-area of circuit conductors (continued)

2.2 determination of conductor size for unburied circuits (continued)

When cables are installed in more than onelayer a further factor, by which K2 must bemultiplied, will have the following values :2 layers : 0.803 layers : 0.734 or 5 layers : 0.70.

Correction factor K3Factor K3 is a measure of the influence of thetemperature, according to the type ofinsulation.

ambient insulationtemperatures elastomer polyvinylchloride cross-linked-

(rubber) (PVC) polyethylene (XLPE)butyl, ethylene-propylene-rubber (EPR)

10 1.29 1.22 1.1515 1.22 1.17 1.1220 1.15 1.12 1.0825 1.07 1.07 1.0430 1.00 1.00 1.0035 0.93 0.93 0.9640 0.82 0.87 0.9145 0.71 0.79 0.8750 0.58 0.71 0.8255 - 0.61 0.7660 - 0.50 0.7165 - - 0.6570 - - 0.5875 - - -80 - - -

factor K3 is a measure of theinfluence of the temperatureaccording to the type of insulation.

table H1-15: correction factor K3 for ambient temperatures other than 30 °C.

Example:A 3-phase 3-core XLPE cable is laid on aperforated cable-tray in close proximity tothree other circuits, consisting of:c a 3-phase 3-core cable (circuit no. 1),c three single-core cables (circuit no. 2),c six single-core cables (circuit no. 3),circuit no. 2 and no. 3 are 3-phase circuits,the latter comprising 2 cables per phase.There are, therefore, effectively 5 3-phasecircuits to be considered, as shown in figureH1-16. The ambient temperature is 40 °C.The code letter indicated in table H1-12 is E.K1 given by table H1-13 = 1.K2 given by table H1-14 = 0.75.K3 given by table H1-15 = 0.91.K = K1 x K2 x K3 = 1 x 0.75 x 0.91 = 0.68.

1 2 3

θa = 40°C XLPE

fig. H1-16: example in the determination offactors K1, K2 and K3.


H1

insulation and number of conductors (2 or 3)rubber butyl or XLPE or EPRor PVC

code B PVC3 PVC2 PR3 PR2 B codeletter C PVC3 PVC2 PR3 PR2 C letter

E PVC3 PVC2 PR3 PR2 EF PVC3 PVC2 PR3 PR2 F

c.s.a. 1.5 15.5 17.5 18.5 19.5 22 23 24 26 1.5 c.s.a.copper 2.5 21 24 25 27 30 31 33 36 2.5 copper(mm2) 4 28 32 34 36 40 42 45 49 4 (mm2)

6 36 41 43 48 51 54 58 63 610 50 57 60 63 70 75 80 86 1016 68 76 80 85 94 100 107 115 1625 89 96 101 112 119 127 138 149 161 2535 110 119 126 138 147 158 169 185 200 3550 134 144 153 168 179 192 207 225 242 5070 171 184 196 213 229 246 268 289 310 7095 207 223 238 258 278 298 328 352 377 95120 239 259 276 299 322 346 382 410 437 120150 299 319 344 371 395 441 473 504 150185 341 364 392 424 450 506 542 575 185240 403 430 461 500 538 599 641 679 240300 464 497 530 576 621 693 741 783 300400 656 754 825 940 400500 749 868 946 1083 500630 855 1005 1088 1254 630

c.s.a. 2.5 16.5 18.5 19.5 21 23 25 26 28 2.5 c.s.a.aluminium 4 22 25 26 28 31 33 35 38 4 alu(mm2) 6 28 32 33 36 39 43 45 49 6 (mm2)

10 39 44 46 49 54 59 62 67 1016 53 59 61 66 73 79 84 91 1625 70 73 78 83 90 98 101 108 121 2535 86 90 96 103 112 122 126 135 150 3550 104 110 117 125 136 149 154 164 184 5070 133 140 150 160 174 192 198 211 237 7095 161 170 183 195 211 235 241 257 289 95120 186 197 212 226 245 273 280 300 337 120150 227 245 261 283 316 324 346 389 150185 259 280 298 323 363 371 397 447 185240 305 330 352 382 430 439 470 530 240300 351 381 406 440 497 508 543 613 300400 526 600 663 740 400500 610 694 770 856 500630 711 808 899 996 630

determination of the minimumcross-sectional area of aconductorThe current Iz when divided by K gives afictitious current I'z. Values of I'z are given intable H1-17 below, together withcorresponding cable sizes for different typesof insulation and core material (copper oraluminium).

table H1-17: case of an unburied circuit: determination of the minimum cable size (c.s.a.),derived from the code letter; conductor material; insulation material and the fictitiouscurrent I'z.


H1


2.2 determination of conductor size for unburied circuits (continued)

ExampleThe example shown in figure H1-16 fordetermining the value of K, will also be usedto illustrate the way in which the minimumcross-sectional-area (c.s.a.) of conductorsmay be found, by using the table H1-17.The XLPE cable to be installed will carry23 amps per phase.Previous examples show that:c the appropriate code letter is E,c the factor K = 0.68.

1 2 3

θa = 40°C XLPE

fig. H1-18: example for the determinationof minimum cable sizes.

Determination of the cross-sectional areasA standard value of In nearest to, but higherthan 23 A is required.Two solutions are possible, one based onprotection by a circuit breaker and the secondon protection by fuses.c circuit breaker: In = 25 Av permissible current Iz = 25 Av fictitious current

I'z = 25 = 36.8 A 0.68v cross-sectional-area of conductors is foundas follows:In the column PR3 corresponding to codeletter E the value of 42 A (the nearest valuegreater than 36.8 A) is shown to require acopper conductor c.s.a. of 4 mm2.For an aluminium conductor thecorresponding values are 43 A and 6 mm2.c fuses: In = 25 Av permissible current Iz = K3In = 1.21 x 25 = Iz = 30.3 A

v the fictitious current I'z = 30.3 = 40.6 A 0.68

v the cross-sectional-areas, of copper oraluminium conductors are (in this case) foundto be the same as those noted above for acircuit-breaker-protected circuit.

2.3 determination of conductor size for buried circuits

In the case of buried circuits thedetermination of minimum conductor sizes,necessitates the establishement of a factor K.

A code letter corresponding to a method ofinstallation is not necessary.

determination of factor KFactor K summarizes the global influence ofdifferent conditions of installation, and isobtained by multiplying together correctionfactors K4, K5, K6 and K7.The values of these several factors are givenin tables H1-19 to H1-22.Correction factor K4Factor K4 is a measure of the influence of themethod of installation.

for buried circuits the value of factorK characteristizes the conditions ofinstallation, and is obtained from thefollowing factors:K4 x K5 x K6 x K7 = Keach of which depends on aparticular feature of installation.

method of installation K4placed in earthenware ducts; in 0.8conduits, or in decorative mouldingsother cases 1table H1-19: correction factor K4 related tothe method of installation.

factor K4 measures the influence ofthe method of installation.

Correction factor K5Factor K5 is a measure of the mutualinfluence of circuits placed side-by-side inclose proximity.Cables are in close proximity when thedistance L separating them is less thandouble the diameter of the larger of the twocables concerned.

factor K5 measures the mutualinfluence of circuits placed side-by-side in close proximity.

location of correction factor K5cables side-by-side number of circuits or of multicore cablesin close proximity 1 2 3 4 5 6 7 8 9 12 16 20buried 1.00 0.80 0.70 0.65 0.60 0.57 0.54 0.52 0.50 0.45 0.41 0.38table H1-20: correction factor K5 for the grouping of several circuits in one layer.

When cables are laid in several layers,multiply K5 by 0.8 for 2 layers, 0.73 for3 layers, 0.7 for 4 layers or 5 layers.


H1

factor K6 is a measure of theinfluence of the earth in which thecable is buried.

Correction factor K6This factor takes into account the nature andcondition of the soil in which a cable (orcables) is (are) buried; notably its thermalconductivity.

nature of soil K6very wet soil (saturated) 1.21wet soil 1.13damp soil 1.05dry soil 1.00very dry soil (sunbaked) 0.86table H1-21: correction factor K6 for thenature of the soil.

factor K7 is a measure of theinfluence of the soil temperature.

Correction factor K7This factor takes into account the influence ofsoil temperature if it differs from 20 °C.

soil temperature insulation°C polyvinyl-chloride cross-linked polyethylene

(PVC) (XLPE)ethylene-propylenerubber (EPR)

10 1.10 1.0715 1.05 1.0420 1.00 1.0025 0.95 0.9630 0.89 0.9335 0.84 0.8940 0.77 0.8545 0.71 0.8050 0.63 0.7655 0.55 0.7160 0.45 0.65table H1-22: correction factor K7 for soil temperatures different than 20 °C.

ExampleA single-phase 230 V circuit is included withfour other loaded circuits in a buried conduit.The soil temperature is 20 °C. Theconductors are PVC insulated and supply a5 kW lighting load. The circuit is protected bya circuit breaker.

K4 from table H1-19 = 0.8.K5 from table H1- 20 = 0.6.K6 from table H1- 21 = 1.0.K7 from table H1- 22 = 1.0.K = K4 x K5 x K6 x K7 = 0.48.

5 kW230 V

θa = 20°C

fig. H1-23: example for the determinationof K4, K5, K6 and K7.


H1


2.3 determination of conductor size for buried circuits (continued)

determination of the smallest

c.s.a. (cross-sectional-area) of a

conductor, for buried circuits

Knowing Iz and K, the corresponding cross-sectional-areas are given in table H1-24below.

insulation and number of loaded conductorsrubber or PVC Butyl, or cross-linked polyethylene

XLPE, or ethylene-propylene rubberEPR

3 conductors 2 conductors 3 conductors 2 conductorsc.s.a. 1.5 26 32 31 37copper 2.5 34 42 41 48(mm2) 4 44 54 53 63

6 56 67 66 8010 74 90 87 10416 96 116 113 13625 123 148 144 17335 147 178 174 20850 174 211 206 24770 216 261 254 30495 256 308 301 360120 290 351 343 410150 328 397 387 463185 367 445 434 518240 424 514 501 598300 480 581 565 677

c.s.a. 10 57 68 67 80aluminium 16 74 88 87 104(mm2) 25 94 114 111 133

35 114 137 134 16050 134 161 160 18870 167 200 197 23395 197 237 234 275120 224 270 266 314150 254 304 300 359185 285 343 337 398240 328 396 388 458300 371 447 440 520

table H1-24: case of a buried circuit: minimum c.s.a. in terms of type of conductor; type ofinsulation; and value of fictitious current I'z (I'z = Iz).

K

ExampleThis is a continuation of the previousexample, for which the factors K4, K5, K6and K7 were determined, and the factor Kwas found to be 0.48.

Full load current

IB = 5.000 = 22 A 230

Selection of protectionA circuit-breaker rated at 25 A would beappropriate.Maximum permanent current permittedIz = 25 A (i.e. the circuit-breaker rating In)

Fictitious current

I'z = Iz = 25 = 52.1 A K 0.48

C.s.a. of circuit conductorsIn the column PVC, 2 conductors, a current of54 A corresponds to a 4 mm2 copperconductor.In the case where the circuit conductors arein aluminium, the same fictitious current(52 A) would require the choice of 10 mm2

corresponding to a fictitious current value(for aluminium) of 68 A.

5 kW230 V

θa = 20°C

fig. H1-25: example for determination ofthe minimum c.s.a. of the circuitconductors.


H1

3. determination of voltage drop

The impedance of circuit conductors is lowbut not negligible: when carrying load currentthere is a fall in voltage between the origin ofthe circuit and the load terminals. The correctoperation of an item of load (a motor; lightingcircuit; etc.) depends on the voltage at itsterminals being maintained at a value close toits rated value. It is necessary therefore todimension the circuit conductors such, that atfull-load current, the load terminal voltage is

maintained within the limits required forcorrect performance.This section deals with methods ofdetermining voltage drops, in order to checkthat:c they conform to the particular standardsand regulations in force,c they can be tolerated by the load,c they satisfy the essential operationalrequirements.

3.1 maximum voltage-drop limit

Maximum allowable voltage-drop limits varyfrom one country to another. Typical valuesfor LV installations are given below in tableH1-26.

maximum voltage-drop between the service-connection point and the point of utilizationlighting other uses

(heating and power)a low-voltage service connection from a LV 3 % 5 %public power distribution networkconsumers HV/LV substation supplied from 6 % 8 %a public distribution HV system

table H1-26: maximum voltage-drop limits.

These voltage-drop limits refer to normalsteady-state operating conditions and do notapply at times of motor starting; simultaneousswitching (by chance) of several loads, etc.as mentioned in Chapter B Sub-clause 4.3(factor of simultaneity, etc.).When voltage drops exceed the valuesshown in table H1-26 larger cables (wires)must be used to correct the condition.The value of 8%, while permitted, can lead toproblems for motor loads; for example:c in general, satisfactory motor performancerequires a voltage within ± 5% of its ratednominal value in steady-state operation,c starting current of a motor can be 5 to7 times its full-load value (or even higher). If8% voltage drop occurs at full-load current,then a drop of 40% or more will occur duringstart-up. In such conditions the motor willeither:v stall (i.e. remain stationary due toinsufficient torque to overcome the loadtorque) with consequent over-heating andeventual trip-out,v or accelerate very slowly, so that the heavycurrent loading (with possibly undesirablelow-voltage effects on other equipment) willcontinue beyond the normal start-up period;c finally an 8% voltage drop represents acontinuous (E2/R watts) of power loss, which,for continuous loads will be a significantwaste of (metered) energy. For these reasonsit is recommended that the maximum value of

8% in steady operating conditions should notbe reached on circuits which are sensitive tounder-voltage problems.

Important:In a number of countries the existing220/380 V 3-phase systems are beinguprated to operate eventually at nominal230/400 V (the recommended IEC standard).Transformer manufacturers in these countrieshave recently increased the no-loadsecondary voltage of their distributiontransformers accordingly, to 237/410 V.After several years of transition in theappliances industry, distribution transformerswill be manufactured with no-load ratios of242/420 V.The rated voltage of consumer appliances willevolve in the same time-scale.From now on, therefore, voltage-dropcalculations must take account of thesechanges.Dangerous possible consequences formotors are:c a lightly-loaded "new" transformer and an"old" motor: risk of overvoltage on the motor,c a fully-loaded "old" transformer and a "new"motor: risk of undervoltage at the motor.Similar (but inverse) problems will arise incountries which presently operate 240/415 Vsystems, if the IEC 230/400 V standard isadopted by them.

load

LV consumer

5%(1)

8%(1)

HV consumer

(1) between the LVsupply point andthe load

fig. H1-27: maximum voltage drop.


H1

circuit voltage drop ( ∆ U)in volts in %

single phase: phase/phase ∆U = 2 IB (R cos ϕ + X sin ϕ) L 100 ∆U Un

single phase: phase/neutral ∆U = 2 IB (R cos ϕ + X sin ϕ) L 100 ∆U Vn

balanced 3-phase: 3 phases ∆U = eIB (R cos ϕ + X sin ϕ) L 100 ∆U(with or without neutral) Un

3. determination of voltage drop (continued)

3.2 calculation of voltage drops in steady load conditions

use of formulae

The table below gives formulae commonlyused to calculate voltage drop in a givencircuit per kilometre of length.If:IB: the full load current in ampsL: length of the cable in kilometresR: resistance of the cable conductor in Ω/km

R = 22,5 Ω.mm2/km for copper S (c.s.a. in mm2)

R = 36 Ω.mm2/km for aluminium S (c.s.a. in mm2)Note: R is negligible above a c.s.a. of 500 mm2.

X: inductive reactance of a conductor in Ω/kmNote: X is negligible for conductors of c.s.a. less than50 mm2. In the absence of any other information, take X asbeing equal to 0.08 Ω/km.

ϕ: phase angle between voltage and currentin the circuit considered, generally:c lighting: cos ϕ = 1c motor power:v at start-up: cos ϕ = 0.35v in normal service: cos ϕ = 0.8Un: phase-to-phase voltage.Vn: phase-to-neutral voltage.For prefabricated pre-wired ducts and bus-trunking, resistance and inductive reactancevalues are given by the manufacturer.

table H1-28: voltage-drop formulae.

simplified table

Calculations may be avoided by using thetable H1-29 below, which gives, with anadequate approximation, the phase-to-phasevoltage drop per km of cable per ampere, interms of:c kinds of circuit use: motor circuits withcos ϕ close to 0.8, or lighting with a cos ϕ inthe neighbourhood of unity;c of the type of cable; single-phase or3-phase.Voltage drop in a cable is then given by:K x IB x L

K is given by the table,IB is the full-load current in amps,L is the length of cable in km.The column motor power cos ϕ = 0.35" oftable H1-29 may be used to compute thevoltage drop occurring during the start-upperiod of a motor (see example (1) after thetable H1-29).

c.s.a. in mm 2 single-phase circuit balanced three-phase circuitmotor power lighting motor power lightingnormal service start-up normal service start-up

Cu Al cos ϕ = 0.8 cos ϕ = 0.35 cos ϕ = 1 cos ϕ = 0.8 cos ϕ = 0.35 cos ϕ = 11.5 24 10.6 30 20 9.4 252.5 14.4 6.4 18 12 5.7 154 9.1 4.1 11.2 8 3.6 9.56 10 6.1 2.9 7.5 5.3 2.5 6.210 16 3.7 1.7 4.5 3.2 1.5 3.616 25 2.36 1.15 2.8 2.05 1 2.425 35 1.5 0.75 1.8 1.3 0.65 1.535 50 1.15 0.6 1.29 1 0.52 1.150 70 0.86 0.47 0.95 0.75 0.41 0.7770 120 0.64 0.37 0.64 0.56 0.32 0.5595 150 0.48 0.30 0.47 0.42 0.26 0.4120 185 0.39 0.26 0.37 0.34 0.23 0.31150 240 0.33 0.24 0.30 0.29 0.21 0.27185 300 0.29 0.22 0.24 0.25 0.19 0.2240 400 0.24 0.2 0.19 0.21 0.17 0.16300 500 0.21 0.19 0.15 0.18 0.16 0.13

table H1-29: phase-to-phase voltage drop ∆U for a circuit, in volts per ampere per km.


H1

examples:

Example 1 (figure H1-30)A three-phase 35 mm2 copper cable50 metres long supplies a 400 V motortaking:c 100 A at a cos ϕ = 0.8 on normalpermanent load;c 500 A (5 In) at a cos ϕ = 0.35 duringstart-up.The voltage drop at the origin of the motorcable in normal circumstances (i.e. with thedistribution board of figure H1-30 distributinga total of 1.000 A) is 10 V phase-to-phase.What is the volt drop at the motor terminals:c in normal service?c during start-up?Solution:c voltage drop in normal service conditions:∆U % = 100 ∆U/UnTable H1-29 shows 1 V/A/km so that:∆U for the cable = 1 x 100 x 0.05 = 5 V∆U total = 10 + 5 = 15 V = i.e. 15 x 100 = 3.75 %400This value is less than that authorized (8%)and is satisfactory.c voltage drop during motor start-up:∆U cable = 0.52 x 500 x 0.05 = 13 VOwing to the additional current taken by themotor when starting, the volt drop at thedistribution board will exceed 10 Volts.Supposing that the infeed to the distributionboard during motor starting is 900 + 500 =1.400 A then the volt-drop at the distributionboard will increase approximately pro rata,i.e.10 x 1400 = 14 V 1000∆U distribution board = 14 V∆U for the motor cable = 13 V∆U total = 13 + 14 = 27 V i.e.27 x 100 = 6.75 % 400a value which is satisfactory during motorstarting.

1000 A

400 V

50 m / 35 mm2 CuIB = 100 A(500 A duringstart-up)

fig. H1-30: example 1.

50 m / 70 mm2 CuIB = 150 A

20 m / 2.5 mm2 CuIB = 20 A


Example 2A 3-phase 4-wire copper line of 70 mm2 c.s.a.and a length of 50 m passes a current of150 A. The line supplies, among other loads,3 single-phase lighting circuits, each of2.5 mm2 c.s.a. copper 20 m long, and eachpassing 20 A.It is assumed that the currents in the 70 mm2

line are balanced and that the three lightingcircuits are all connected to it at the samepoint.What is the voltage drop at the end of thelighting circuits?

Solution:c voltage drop in the 4-wire line:∆U % = 100 ∆U/UnTable H1-29 shows 0.55 V/A/km.∆U line = 0.55 x 150 x 0.05 = 4.125 Vphase-to-phase

which: 4.125 V = 2.38 V phase to neutral. ec voltage drop in any one of the lightingsingle-phase circuits:∆U for a single-phase circuit = 18 x 20 x 0.02= 7.2 VThe total volt-drop is therefore7.2 + 2.38 = 9.6 V 9.6 V x 100 = 4.2 %230 VThis value is satisfactory, being less than themaximum permitted voltage drop of 6%.


H1

4. short-circuit current calculations

knowing the levels of 3-phasesymmetrical short-circuit currents(Isc) at different points in aninstallation is an essential feature ofits design.

A knowledge of 3-phase symmetrical short-circuit current values (Isc) at strategic pointsof an installation is necessary in order todimension switchgear (fault current rating);cables (thermal withstand rating); protectivedevices (discriminative trip settings) and soon...In the following notes a 3-phase short-circuitof zero impedance (the so-called boltedshort-circuit) fed through a typical HV/LVdistribution transformer will be examined.

Except in very unusual circumstances, thistype of fault is the most severe, and iscertainly the simplest to calculate.Short-circuit currents occurring in a networksupplied from an alternator and also in d.c.systems are dealt with in Chapter JSub-clauses 1.1 and 6.1.The simplified calculations and practical ruleswhich follow give conservative results ofsufficient accuracy, in the large majority ofcases, for installation design purposes.

4.1 short-circuit current at the secondary terminals of a HV/LV distribution transformer

the case of one transformer

c as a first approximation the impedance ofthe HV system is assumed to be negligiblysmall, so that:

Isc = In x 100 where Usc

In = P x 103 and: eU20P = kVA rating of the transformer,U20 = phase-to-phase secondary volts onopen circuit,In = nominal current in ampsIsc = short-circuit fault current in amps,Usc = short-circuit impedance voltage of thetransformer in %.Typical values of Usc for distributiontransformers are given in the table H1-32.

Example:400 kVA transformer, 242/420 V at no loadUsc = 4%

In = 400 x 103 = 550 A ÷ ex 420

Isc = 550 x 100 = 13.75 kA 4c in practice Isc is slightly less than thatcalculated by this method, as shown in thefollowing table (H1-33) since the HV systemimpedance is such that its fault level at theHV terminals of the transformer rarelyexceeds 500 MVA. A level of 250 MVA, orless, is more common.

transformer Usc in %rating type of transformer

oil-immersed cast-resin50 to 630 4% 6%800 to 2500 6% 6%

table H1-32: typical values of Usc fordifferent kVA ratings of transformers withHV windings iiiii 20 kV.

the case of several transformers

in parallel feeding a busbar

The value of fault current on an outgoingcircuit immediately downstream of thebusbars (figure H1-34) can be estimated asthe sum of the Isc from each transformercalculated separately.It is assumed that all transformers aresupplied from the same HV network, in whichcase the values obtained from table H1-33when added together will give a slightlyhigher fault-level value than would actuallyoccur.Other factors which have not been taken intoaccount are the impedance of the busbarsand of the circuit breakers.The conservative fault-current value obtainedhowever, is sufficiently accurate for basicinstallation design purposes.

tables H1-33: Isc at the LV terminals of 3-phase HV/LV transformers supplied from a HV system with a 3-phase fault level of 500 MVA, or250 MVA.

The choice of circuit breakers andincorporated protective devices againstshort-circuit fault currents is described inChapter H2 Sub-clause 4.4.

Isc1 Isc3Isc2

Isc1 + Isc2 + Isc3

fig. H1-34.

transformer rated power (kVA) 50 100 160 250 315 400 500 630 800 1000 1250 1600 2000 2500transformer current Ir (A) 69 137 220 344 433 550 687 866 1100 1375 1718 2199 2749 3437oil-immersed transformer Isc (kA) Psc = 250 MVA 1.71 3.40 5.41 8.38 10.5 13.2 16.4 20.4 17.4 21.5 26.4 33.1 40.4 49.1

Psc = 500 MVA 1.71 3.42 5.45 8.49 10.7 13.5 16.8 21.0 17.9 22.2 27.5 34.8 43.0 52.9cast-resin transformer Isc (kA) Psc = 250 MVA 1.14 2.28 3.63 5.63 7.07 8.93 11.1 13.9 17.4 21.5 26.4 33.1 40.4 49.1

Psc = 500 MVA 1.14 2.28 3.65 5.68 7.14 9.04 11.3 14.1 17.9 22.2 27.5 34.8 43.0 52.9


H1

4.2 3-phase short-circuit current (Isc) at any point within a LV installation

In a 3-phase installation Isc at any point isgiven by:

Isc = U20 (amps) eZTU20 = phase-to-phase voltage of the open-circuited secondary windings of the power-supply transformer(s).ZT = total impedance per phase of theinstallation upstream of the fault location(in ohms).

method of calculating ZT

Each component of an installation (HVnetwork, transformer, cable, circuit breaker,busbar, and so on...) is characterized by itsimpedance Z, comprising an element ofresistance (R) and an inductive reactance(X). It may be noted that capacitivereactances are not important in short-circuitcurrent calculations.The parameters R, X and Z are expressed inohms, and are related by the sides of a right-angled triangle, as shown in the impedancediagram of figure H1-35.The method consists in dividing the networkinto convenient sections, and to calculate theR and X values for each.Where sections are connected in series in thenetwork all the resistive elements in thesection are added arithmetically; likewise forthe reactances, to give RT and XT. Theimpedance (Z) for the combined sectionsconcerned is then calculated from

Z R XT TT = +2 2

Any two sections of the network which areconnected in parallel, can, if, predominantlyboth resistive (or both inductive) be combinedto give a single equivalent resistance (orreactance) as follows:Let R1 and R2 be the two resistancesconnected in parallel, then the equivalentresistance R3 wil be given by:

R3 = R1 R2 or for reactances X3 = X1 X2 R1 + R2 X1 + X2Combining two or more dissimilar circuits inparallel is (fortunately) seldom required innormal radial-type installation networks andwill not be demonstrated in the main text.General methods for reducing impedances toa single equivalent impedance are given,however, in Appendix H1.

ZX

Rϕ

fig. H1-35: impedance diagram.

determination of the impedance

of the HV network

c network upstream of the HV/LVtransformer (table H1-36)The 3-phase short-circuit fault level in kA or inMVA* is given by the power supply authorityconcerned, from which an equivalentimpedance can be deduced.A formula which makes this deduction and atthe same time converts the impedance to anequivalent value at LV is given, as follows

Zs = Uo2 where: PscZs = impedance of the HV voltage network,expessed in milli-ohms,Uo = phase-to-phase no-load LV voltage,expressed in volts.

Psc = HV 3-phase short-circuit fault level,expressed in kVA.*Short-circuit MVA: eEL Isc where:EL = phase-to-phase nominal system voltage expressed inkV (r.m.s.).Isc = 3-phase short-circuit current expressed in kA (r.m.s.).

The upstream (HV) resistance Ra is generallyfound to be negligible compared with thecorresponding Xa, the latter then being takenas the ohmic value for Za. If more accuratecalculations are necessary, Ra may be takento be equal to 0.15 Xa.Table H1-36 gives values for Ra and Xacorresponding to the most common HV*short-circuit levels in public power-supplynetworks, namely, 250 MVA and 500 MVA.* up to 36 kV

Psc Uo (V) Ra (m Ω) Xa (mΩ)250 MVA 420 0.106 0.71500 MVA 420 0.053 0.353

table H1-36: the impedance of the HV network referred to the LV side of the HV/LVtransformer.


H1

4. short-circuit current calculations (continued)

4.2. 3-phase short-circuit current (Isc) at any point within a LV installation (continued)

c transformers (table H1-37)The impedance Ztr of a transformer, viewedfrom the LV terminals, is given by the formula:

Ztr = U2

2O x Usc

milli-ohms where: Pn 100U2 o = open-circuit secondary phase-to-phase voltage expressed in volts.Pn = rating of the transformer (in kVA).Usc = the short-circuit impedance voltage ofthe transformer expressed in %.The transformer windings resistance Rtr canbe derived from the total losses as follows:

Pcu = 3 In2 Rtr so that Rtr = Pcu x 103

3 In2

in milli-ohms wherePcu = total losses in watts.In = nominal full-load current in amps.Rtr = resistance of one phase of thetransformer in milli-ohms (the LV andcorresponding HV winding for one LV phaseare included in this resistance value).

For an approximate calculation Rtr may beignored since X ≈ Z in standard distribution-type transformers.

table H1-37: resistance, reactance and impedance values for typical distribution transformers with HV windings iiiii 20 kV.

c circuit breakersIn LV circuits, the impedance of circuitbreakers upstream of the fault location mustbe taken into account. The reactance valueconventionally assumed is 0.15 mΩ per CB,while the resistance is neglected.c busbarsThe resistance of busbars is generallynegligible, so that the impedance ispractically all reactive, and amounts toapproximately 0.15 mΩ/metre* length for LVbusbars (doubling the spacing between thebars increases the reactance by about 10%only).* for 50 Hz systems, but 0.18 mΩ/metre length at 60 Hz.

c circuit conductorsThe resistance of a conductor is given by theformula:

Rc = ρ x L where Sρ = the resistivity constant of the conductormaterial at the normal operating temperaturebeing:22.5 mΩ.mm2/m for copper,36 mΩ.mm2/m for aluminium,S = c.s.a. of conductor in mm2.Cable reactance values can be obtained fromthe manufacturers. For c.s.a. of less than50 mm2 reactance may be ignored. In theabsence of other information, a value of0.08 mΩ/metre may be used (for 50 Hzsystems) or 0.096 mΩ/metre (for 60 Hzsystems). For prefabricated bus-trunking andsimilar pre-wired ducting systems, themanufacturer should be consulted.

c motorsAt the instant of short-circuit, a running motorwill act (for a brief period) as a generator, andfeed current into the fault.In general, this fault-current contribution maybe ignored. However, for more precisecalculation, particularly in the case of largemotors and/or numerous smaller motors, thetotal contribution can be estimated from theformula:Iscm = 3.5 In from each motori.e. 3.5 m In for m similar motors operatingconcurrently.The motors concerned will be the 3-phasemotors only; single-phase-motor contributionbeing insignificant.For HV circuit breakers, the contribution frommotors is often reduced to very low values atthe instant of contact separation, and for thatreason is frequently ignored, but with low-inertia high-speed LV CBs* the value given inthe above formula is recommended.Note: for circuit breaker fault-current making-duty however, no account can be taken ofdiminution of fault current contribution, andeach running motor will initially feed currentinto the fault at a level approaching its ownstarting current, i.e. 4 or 5 In.* and fuses

c fault-arc resistanceShort-circuit faults generally form an arcwhich has the properties of resistance. Theresistance is not stable and its average valueis low, but at low voltage this resistance issufficient to reduce the fault-current to someextent. Experience has shown that areduction of the order of 20% may beexpected. This phenomenon will effectivelyease the current-breaking duty of a CB, butaffords no relief for its fault-current makingduty.

transformer rated power kVA 50 100 160 250 315 400 500 630 800 1000 1250 1600 2000 2500oil-immersed transformer Usc % 4 4 4 4 4 4 4 4 6 6 6 6 6 6

Rtr mΩ 95.3 37.9 16.2 9.2 6.9 5.1 3.9 2.9 2.9 2.3 1.8 1.4 1.1 0.9Xtr mΩ 104.1 59.5 41.0 26.7 21.3 16.9 13.6 10.8 12.9 10.3 8.3 6.5 5.2 4.1Ztr mΩ 141.1 70.5 44.1 28.2 22.4 17.7 14.1 11.2 13.2 10.6 8.5 6.6 5.3 4.2

cast-resin transformer Usc % 6 6 6 6 6 6 6 6 6 6 6 6 6Rtr mΩ 33.5 18.6 10.7 8.2 6.1 4.6 3.5 2.6 1.9 1.5 1.1 0.8 0.6Xtr mΩ 100.4 63.5 41.0 32.6 25.8 20.7 16.4 13.0 10.4 8.3 6.5 5.2 4.2Ztr mΩ 105.8 66.2 42.4 33.6 26.5 21.2 16.8 13.3 10.6 8.4 6.6 5.3 4.2

Xtr Ztr Rtr= −2 2


H1

system elements considered resistance R in milli-ohms reactance X in milli-ohmssupply network Ra ≈ 0,15 Xa ≈ Za =

U22 0

table H1-33 Xa PscR can therefore be neglected in comparison with X

transformer Rtr = Pcu x 103

table H1-34 3In2

Rtr = is often negligible compared to Xtr avec Ztr = U22 0 x

Uscfor transformers > 100 kVA Pn 100

circuit breaker negligible XD = 0.15 mΩ/polebusbars negligible for S > 200 mm2 in the formula XB = 0.15 mΩ/m

below:

R = ρ L (1) S

circuit conductors (2) R = ρ L (1) cables : Xc = 0.08 mΩ/m S

motors see Sub-clause 4.2 "motors"(often negligible at LV))

three-phase short-circuit current in kA

Ztr Rtr−2 2

recapitulation table

M

3=Isc

RT2 + XT2

U2o

table H1-38: recapitulation table of impedances for different parts of a power-supply system.

U20: phase-to-phase no-load secondaryvoltage of HV/LV transformer (in volts).Psc: 3-phase short-circuit power at HVterminals of the HV/LV transformers (in kVA).Pcu: 3-phase total losses of the HV/LVtransformer (in watts).Pn: rating of the HV/LV transformer (in kVA).Usc: short-circuit impedance voltage of theHV/LV transfomer (in %).

RT : total resistance. XT: total reactance(1) ρ = resistivity at normal temperature ofconductors in serviceρ = 22.5 milli-ohms x mm2/metre for copperρ = 36 milli-ohms x mm2/metre for aluminium.(2) If there are several conductors in parallelper phase, then divide the resistance of oneconductor by the number of conductors. Thereactance remains practically unchanged.

R (mΩ) X (mΩ) RT (mΩ) XT (mΩ)

HV network 0.053 0.353Psc = 500 MVAtransformer 2.35 10.3420 kV/420 VPn = 1000 kVAUsc = 6%Pcu = 13.3 x 103 wattssingle-core cables Rc = 22.5 x 5 Xc = 0.08 x 5 2.523 11.15 m copper 4 2404 x 240 mm2/phase = 0.12 = 0.40 Isc1 = 21.3 kAmain RD = 0 XD = 0.15circuit breakerbusbars RB = 0 XB = 1.5 2.523 12.75 Isc2 = 18.6 kA10 mthree-core cable Rc = 22.5 x 100 Xc = 100 x 0.08100 m 9595 mm2 copper = 23.68 = 8 26.2 20.75 Isc3 = 7.24 kAthree-core cable Rc = 22.5 x 20 Xc = 20 x 0.0820 m 1010 mm2 copper = 45 = 1.6 71.2 22.35 Isc4 = 3.24 kAfinal circuits

3=Isc

RT2 + XT2

420

table H1-39: example of short-circuit current calculations for a LV installation supplied at 400 V (nominal) from a 1,000 kVA H V/LVtransformer.

4.3 Isc at the receiving end of a feeder in terms of the Isc at its sending end

The following tables, derived by the "methodof composition" (mentioned in Chapter GSub-clause 5.2) give a rapid and sufficientlyaccurate value of short-circuit current at apoint in a network, knowing:c the value of short-circuit current upstreamof the point consideredc the length and composition of the circuitbetween the point at which the short-circuitcurrent level is known, and the point at whichthe level is to be determined.It is then sufficient to select a circuit breaker withan appropriate short-circuit fault ratingimmediately above that indicated in the tables.

If more precise values are required, it ispossible to make a detailled calculation (seeSub-Clause 4.2 above) or to use a softwarepackage, such as Ecodial*.In such a case, moreover, the possibility ofusing the cascading technique should beconsidered, in which the use of a current-limiting circuit breaker at the upstreamposition would allow all circuit breakersdownstream of the limiter to have a short-circuit-current rating much lower than wouldotherwise be necessary (see Chapter H2Sub-clause 4.5).* a Merlin Gerin product (see Chapter B, Clause 1,Methodology).


H1

4. short-circuit current calculations (continued)

4.3 Isc at the receiving end of a feeder in terms of the Isc at its sending end (continued)

copper c.s.a. of length of circuit (in metres)230 V / phase400 V conductors

(in mm 2)1.5 0.8 1 1.3 1.6 32.5 1 1.3 1.6 2.1 2.6 54 0.8 1.7 2.1 2.5 3.5 4 8.56 1.3 2.5 3 4 5 6.5 1310 0.8 1.1 2.1 4 5.5 6.5 8.5 11 2116 0.9 1 1.4 1.7 3.5 7 8.5 10 14 17 3425 1 1.3 1.6 2.1 2.6 5 10 13 16 21 26 5035 1.5 1.9 2.2 3 3.5 7.5 15 19 22 30 37 7550 1.1 2.1 2.7 3 4 5.5 11 21 27 32 40 55 11070 1.5 3 3.5 4.5 6 7.5 15 30 37 44 60 75 15095 0.9 1 2 4 5 6 8 10 20 40 50 60 80 100 200120 0.9 1 1.1 1.3 2.5 5 6.5 7.5 10 13 25 50 65 75 100 130 250150 0.8 1 1.1 1.2 1.4 2.7 5.5 7 8 11 14 27 55 70 80 110 140 270185 1 1.1 1.3 1.5 1.6 3 6.5 8 9.5 13 16 32 65 80 95 130 160 320240 1.2 1.4 1.6 1.8 2 4 8 10 12 16 20 40 80 100 120 160 200 400300 1.5 1.7 1.9 2.2 2.4 5 9.5 12 15 19 24 49 95 120 150 190 2402 x 120 1.5 1.8 2 2.3 2.5 5.1 10 13 15 20 25 50 100 130 150 200 2502 x 150 1.7 1.9 2.2 2.5 2.8 5.5 11 14 17 22 28 55 110 140 170 220 2802 x 185 2 2.3 2.6 2.9 3.5 6.5 13 16 20 26 33 65 130 160 200 260 3303 x 120 2.3 2.7 3 3.5 4 7.5 15 19 23 30 38 75 150 190 230 300 3803 x 150 2.5 2.9 3.5 3.5 4 8 16 21 25 33 41 80 160 210 250 330 4103 x 185 2.9 3.5 4 4.5 5 9.5 20 24 29 39 49 95 190 240 290 390Isc upstream Isc downstream(in kA) (in Ka)100 94 94 93 92 91 83 71 67 63 56 50 33 20 17 14 11 9 590 85 85 84 83 83 76 66 62 58 52 47 32 20 16 14 11 9 4.580 76 76 75 75 74 69 61 57 54 49 44 31 19 16 14 11 9 4.570 67 67 66 66 65 61 55 52 49 45 41 29 18 16 14 11 5 4.560 58 58 57 57 57 54 48 46 44 41 38 27 18 15 13 10 8.5 4.550 49 48 48 48 48 46 42 40 39 36 33 25 17 14 13 10 8.5 4.540 39 39 39 39 39 37 35 33 32 30 29 22 15 13 12 9.5 8 4.535 34 34 34 34 34 33 31 30 29 27 26 21 15 13 11 9 8 4.530 30 29 29 29 29 28 27 26 25 24 23 19 14 12 11 9 7.5 4.525 25 25 25 24 24 24 23 22 22 21 20 17 13 11 10 8.5 7 420 20 20 20 20 20 19 19 18 18 17 17 14 11 10 9 7.5 6.5 415 15 15 15 15 15 15 14 14 14 13 13 12 9.5 8.5 8 7 6 410 10 10 10 10 10 10 9.5 9.5 9.5 9.5 9 8.5 7 6.5 6.5 5.5 5 3.57 7 7 7 7 7 7 7 7 6.5 6.5 6.5 6 5.5 5 5 4.5 4 2.95 5 5 5 5 5 5 5 5 5 5 5 4.5 4 4 4 3.5 3.5 2.54 4 4 4 4 4 4 4 4 4 4 4 3.5 3.5 3.5 3 3 2.9 2.23 3 3 3 3 3 3 3 3 2.9 2.9 2.9 2.8 2.7 2.6 2.5 2.4 2.3 1.92 2 2 2 2 2 2 2 2 2 2 2 1.9 1.9 1.8 1.8 1.7 1.7 1.41 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.9 0.9 0.9 0.8

aluminium c.s.a. of length of circuit (in metres)230 V / phase400 V conductors

(in mm 2)2.5 0.8 1 1.3 1.6 34 1 1.3 1.6 2.1 2.6 56 0.8 1.6 2 2.4 3 4 810 1.3 2.6 3.5 4 5.5 6.5 1316 0.8 1.1 2.1 4 5.5 6.5 8.5 11 2125 0.8 1 1.3 1.7 3.5 6.5 8.5 10 13 17 3335 0.9 1.2 1.4 1.8 2.3 4.5 9 12 14 18 23 4650 1.3 1.7 2 2.6 3.5 6.5 13 17 20 26 33 6570 0.9 1.8 2.3 2.8 3.5 4.5 9 18 23 28 37 46 9095 1.3 2.5 3 4 5 6.5 13 25 32 38 50 65 130120 0.8 1.7 3 4 4.5 6.5 8 17 32 40 47 65 80 160150 0.9 1.7 3.5 4.5 5 7 8.5 17 34 43 50 70 85 170185 0.9 1 2 4 5 6 8 10 20 40 50 60 80 100 240240 0.9 1 1.1 1.3 2.5 5 6.5 7.5 10 13 25 50 65 75 100 130 250300 0.9 1 1.2 1.4 1.5 3 6 7.5 9 12 15 30 60 75 90 120 150 3002 x 120 0.9 1.1 1.3 1.4 1.6 3 6.5 8 9.5 13 16 32 65 80 95 130 160 3202 x 150 1 1.2 1.4 1.5 1.7 3.5 7 9 10 14 17 35 70 85 100 140 1702 x 185 1.2 1.4 1.6 1.8 2 4.1 8 10 12 16 20 41 80 100 120 160 2002 x 240 1.5 1.8 2 2.3 2.5 5 10 13 15 20 25 50 100 130 150 200 2503 x 120 1.4 1.7 1.9 2.1 2.4 4.5 9.5 12 14 19 24 48 95 120 140 190 2403 x 150 1.5 1.8 2.1 2.3 2.6 5 10 13 15 21 26 50 100 130 150 210 2603 x 185 1.8 2.1 2.4 2.7 3 6 12 15 18 24 30 60 120 150 180 240 3003 x 240 2.3 2.7 3 3.5 4 7.5 15 19 23 30 38 75 150 190 230 300 380

table H1-40: Isc at a point downstream, in terms of a known upstream fault-current value and the length and c.s.a. of the inter veningconductors, in a 230/400 V 3-phase system.

Note: for a 3-phase system having 230 V between phases, divide the above lengths by e= 1.732


H1

Example:The network shown in figure H1-41 typifies acase for the application of table H1-40.Select the c.s.a. of the conductor in thecolumn for copper conductors (in thisexample the c.s.a. is 50 mm2).Search along the row corresponding to50 mm2 for the length of conductor equal tothat of the circuit concerned (or the nearestpossible on the low side). Descend verticallythe column in which the length is located, andstop at a row in the middle section (of the3 sections of the table) corresponding to theknown fault-current level (or the nearest to iton the high side).In this case 30 kA is the nearest to 28 kA onthe high side. The value of short-circuitcurrent at the downstream end of the11 metre circuit is given at the intersection ofthe vertical column in which the length islocated, and the horizontal row correspondingto the upstream Isc (or nearest to it on thehigh side).This value in the example is seen to be19 kA.The procedure for aluminium conductors issimilar, but the vertical column must beascended into the middle section of the table.In consequence, a DIN-rail-mounted circuitbreaker rated at 63 A and Isc of 50 kA (suchas a NC100LH unit*) can be used for the55 A circuit in figure H1-41.A Compact* rated at 260 A with an Isccapacity of 25 kA (such as a NS160N unit*)can be used to protect the 160 A circuit.* Merlin Gerin product.

fig. H1-41: determination of downstreamshort-circuit current level Isc using tableH1-40.

4.4 short-circuit current supplied by an alternator or an inverter

Please refer to Chapter J.

400 VIcc = 28 kA

50 mm2, Cu11 m

Icc = ?

IB = 55 A IB = 160 A


H1

5. particular cases of short-circuit current

5.1 calculation of minimum levels of short-circuit current

if a protective device in a circuit isintended only to protect againstshort-circuit faults, it is essential thatit will operate with certainty at thelowest possible level of short-circuitcurrent that can occur on the circuit.

In general, on LV circuits, a single protectivedevice protects against all levels of current,from the overload threshold through themaximum rated short-circuit current-breakingcapability of the device.

In certain cases, however, overload protectivedevices and separate short-circuit protectivedevices are used.

examples of such arrangements

Figures H1-42 to H1-44 show some commonarrangements where overload and short-circuit protections are effected by separatedevices.As shown in figures H1-42 and H1-43, themost common circuits using separate devicescontrol and protect motors.Figure H1-44 constitutes a derogation in thebasic protection rules, and is generally usedon circuits of prefabricated bustrunking,lighting rails, etc.

aM fuses(no protectionagainst overload)

load-breakingcontactor with thermaloverload relay

fig. H1-42: circuit protected by aM fuses.

circuit breakerwith instantaneousmagnetic short-circuitprotective relay only

load-breakingcontactor with thermaloverload relay

fig. H1-43: circuit protected by circuitbreaker without thermal overload relay.

circuit breaker D

load withincorporatedoverloadprotection

S1

S2 < S1

fig. H1-44: circuit breaker D providesprotection against short-circuit faults asfar as and including the load.


H1

it is necessary that the protectivedevice instantaneous-trip settingc Im < Isc (min)for protection by a circuit breaker;or fusion currentc Ia < Isc (min)for protection by fuses.

conditions to be respected

The protective device must therefore satisfythe two following conditions:c its fault-current breaking rating > Isc the3-phase short-circuit current at its point ofinstallation,c elimination of the minimum short-circuitcurrent possible in the circuit, in a time tccompatible with the thermal constraints of thecircuit conductors, where:

tc = K2 S2 (tc < 5 seconds) Isc (min)Comparison of the tripping or fusingperformance curve of protective devices, withthe limit curves of thermal constraint for aconductor shows that this condition issatisfied if:c Isc (min) > Im (instantaneous or short time-delay circuit-breaker trip setting current level).See figure H1-45.c Isc (min) > Ia for protection by fuses. Thevalue of the current Ia corresponds to thecrossing point of the fuse curve and the cablethermal withstand curve (figures H1-46 andH1-47).

fig. H1-45: protection by circuit breaker.I

t

Im

t =k2 S2

I2

I

t

Ia

t =k2 S2

I2

fig. H1-46: protection by aM-type fuses.

I

t

Ia

t =k2 S2

I2

fig. H1-47: protection by gl-type fuses.

practical method of calculating

Lmax

The limiting effect of the impedance of longcircuit conductors on the value of short-circuitcurrents must be checked and the length of acircuit must be restricted accordingly.The method of calculating the maximumpermitted length has already beendemonstrated in TN- and IT- earthedschemes for single and double earth faults,respectively (see Chapter G Sub-clauses 5.2and 6.2).Two cases are considered below:1. calculation of Lmax for a 3-phase 3-wirecircuitThe minimum short-circuit current will occurwhen two phase wires are short circuited atthe remote end of the circuit.

in practice this means that the lengthof circuit downstream of theprotective device must not exceed acalculated maximum length: Lmax = 0.8 x U x Sph

2 x ρ x Im

load

P

0.8 U L

Using the "conventional method", the voltageat the point of protection P is assumed to be80% of the nominal voltage during a short-circuit fault, so that 0.8 U = Isc Zd, where:Zd = impedance of the fault loopIsc = short-circuit current (ph/ph)U = phase-to-phase nominal voltage.For cables i 120 mm2, reactance may beneglected, so that

Zd = ρ 2L Sph

where ρ = resistivity of copper* at the averagetemperature during a short-circuit, andSph : c.s.a. of a phase conductor in mm2

L = length in metresIn order that the cable will not be damaged byheat Isc u Im

0.8 U u ρ 2L Im or Sph

Lmax = 0.8 U Sph 2 ρ Imwith: U = 400 Vρ = 1.5 x 0.018 = 0.027 Ω.mm2/m**.Im = magnetic trip current setting for the CBLmax = maximum circuit length in metres.

Lmax = 5.926 Sph Im

* or for aluminium according to conductor material** the high value for resistivity is due to the elevatedtemperature of the conductor when passing short-circuitcurrent.


H1

5. particular cases of short-circuit current (continued)

5.1 calculation of minimum levels of short-circuit current (continued)

2. calculation of Lmax for a 3-phase 4-wire230/400 V circuitThe minimum Isc will occur when the short-circuit is between a phase conductor and theneutral.A calculation similar to that of example 1above is required, but using the followingformulae (for cable i 120 mm2 (1)).c where Sn for the neutral conductor = Sphfor the phase conductor

Lmax = 3,421 Sph Im

c If Sn for the neutral conductor < Sph, then

Lmax = 6,842 Sph where m = Sph (1+m) Im Sn(1) for larger c.s.a.'s, the resistancecalculated for the conductors must beincreased to account for the non-uniform

current density in the conductor (due to "skin"and "proximity" effects, previously noted inChapter G Sub-clause 5.2).Suitable values (taken from French standardNF 15-100) are as follows:150 mm2 : R + 15%185 mm2 : R + 20%240 mm2 : R + 25%300 mm2 : R + 30%where R is the value calculated from

R = ρ 2L SphFor larger c.s.a.'s than those listed, reactancevalues must be combined with those ofresistance to give an impedance. Reactancemay be taken as 0.08mΩ/metre for cables(at 50 Hz). At 60Hz the constant is0.096 mΩ/metre.

tabulated values for LmaxTable H1-49 below gives maximum circuitlengths (Lmax) in metres, for:c 3-phase 3-wire 400 V circuits (i.e. withoutneutral) andc 1-phase 2-wire 400 V circuits, withoutneutral, protected by general-purpose circuitbreakers.

In other cases, apply correction factors (givenin table H1-53) to the lengths obtained.

The calculations are based on the foregoingmethods, with Im = 1.2 IrmIrm = regulated short-circuit current tripsetting. Irm is guaranteed to be within ± 20%of the regulated value; hence the worst-casefactor of 1.2 (i.e. 120%).Refer to Chapter H2 Sub-clause 4.2 fordetails of regulation of circuit-breakerprotective elements.

operating current c.s.a. (nominal cross-sectional-area) of conductors (in mm 2)level Im of theinstantaneousmagnetic trippingelement(in A) 1.5 2.5 4 6 10 16 25 35 50 70 95 120 150 185 24050 148 246 39463 117 195 313 47080 92 154 246 370100 74 123 197 296 493125 59 99 158 237 395160 46 77 123 185 308 494200 37 62 99 148 247 395250 30 49 79 118 197 316 494320 23 38 62 92 154 247 386400 18 31 49 74 123 197 308 432500 15 25 39 59 99 158 247 345 494560 13 22 35 53 88 141 220 308 441630 12 19 31 47 78 125 196 274 392700 11 18 28 42 70 113 176 247 353 494800 9 15 25 37 61 98 154 215 308 432875 8 14 22 34 56 90 141 197 282 3951000 7 12 20 30 49 79 123 173 247 345 4691120 6 11 17 26 44 70 110 154 220 308 4191250 6 10 16 24 39 63 99 138 197 276 375 4741600 7 12 18 31 49 77 108 154 216 293 370 5322000 6 10 15 24 39 62 86 123 173 234 296 425 5702500 8 12 20 31 49 69 99 138 188 237 340 438 5923200 6 9 15 25 38 54 77 108 146 185 265 340 4624000 7 12 20 31 43 62 86 117 148 212 273 3705000 6 10 16 25 34 49 69 94 118 170 218 2966300 8 12 20 27 39 55 74 94 134 175 2358000 6 10 15 21 31 43 59 74 105 136 18510000 8 12 17 25 35 47 59 85 109 14712500 6 10 14 20 28 37 47 67 87 118table H1-49: maximum circuit lengths in metres for copper conductors (for aluminium, the lengths must be multiplied by 0.62).


H1

Tables H1-50 to H1-52 below give maximumcircuit length (Lmax) in metres for:c 3-phase 3-wire 400 V circuits (i.e. withoutneutral) andc 1-phase 2-wire 400 V circuits, withoutneutral,protected in both cases by domestic-typecircuit breakers or with circuit breakers havingsimilar tripping/current characteristics.In other cases, apply correction factors to thelengths indicated. These factors are given intable H1-53. The calculations are carried out

according to the method described above,again, with Im = 1.2 Irm as previously noted.These circuit breakers have fixed overload(thermal) tripping elements and fixed short-circuit (magnetic) tripping elements.Categories B, C and D differ only in the levelsof short-circuit-current trip setting Im.IEC 898 is the relevant international standardfor these circuit breakers.See also table H2-28 for tripping ranges.

rated current cross-sectional-area (c.s.a.) of conductors (in mm 2)of circuitbreakers(in A) 1.5 2.5 4 6 10 16 25 35 506 296 494 79010 178 296 474 71113 137 228 385 547 91216 111 185 296 444 74120 89 148 237 356 593 94825 71 119 190 284 474 75932 56 93 148 222 370 593 92640 44 74 119 178 296 474 74150 36 59 95 142 237 379 593 83063 28 47 75 113 188 301 470 658 85480 22 37 59 89 148 237 370 519 704100 18 30 47 71 119 190 296 415 563125 14 24 38 57 95 152 237 331 450table H1-50: maximum length of copper-conductored circuits in metres protected byB-type circuit breakers.

rated current cross-sectional-area (c.s.a.) of conductors (in mm 2)of circuitbreakers(in A) 1.5 2.5 4 6 10 16 25 35 506 148 247 395 593 98810 89 148 237 356 593 94813 68 114 182 274 456 72916 56 93 148 222 370 593 92620 44 74 119 178 296 474 74125 36 59 95 142 237 379 593 83032 28 46 74 111 185 296 463 648 88040 22 37 59 89 148 237 370 519 70450 18 30 47 71 119 190 296 415 56363 14 24 38 56 94 150 235 329 44680 11 19 30 44 74 119 185 259 351100 9 15 24 36 59 95 148 207 281125 7 12 19 28 47 76 119 166 225table H1-51: maximum length of copper-conductored circuits in metres protected byC-type circuit breakers.

rated current cross-sectional-area (c.s.a.) of conductors (in mm 2)of circuitbreakers(in A) 1.5 2.5 4 6 10 16 25 35 506 105 176 283 423 706 112910 63 105 170 254 423 639 105813 48 81 130 195 325 521 814 114016 40 65 105 158 264 422 661 925 125520 32 52 84 126 211 337 528 740 100425 25 41 67 101 169 270 423 592 80332 20 32 52 79 132 211 330 462 62740 16 26 42 63 105 168 264 370 50250 12 20 33 50 84 135 211 296 40163 10 16 26 40 67 107 167 234 31880 8 13 21 31 52 84 132 185 251100 6 10 16 25 42 67 105 148 200125 5 8 13 20 33 54 84 118 160table H1-52: maximum length of copper-conductored circuits in metres protected byD-type circuit breakers (Merlin Gerin).

Note: IEC 898 provides for an upper short-circuit-current tripping range of 10-50 In fortype D circuit breakers. European standards,and the above table H1-52 however, are

based on a range of 10-20 In, a range whichcovers the vast majority of domestic andsimilar installations.


H1

5. particular cases of short-circuit current (continued)

5.1 calculation of minimum levels of short-circuit current (continued)

circuit details Sph = 1 Sph = 2S neutral S neutral

3-phase 3-wire or 1 ph 2-wire 1400 V circuit 400V circuit (no neutral)3-phase 4-wire or 2 ph 3-wire 0.58 0.39 (1)230/400 V circuit 230/400 V circuit

(i.e. with neutral)1-phase 2-wire 0.58(phase and neutral)230 V circuittable H1-53: correction factors to apply to lengths obtained from tables H1-49 to H1-52.

(1) 0.77 for the c.s.a. of the neutral conductor.

examplesExample 1In a 3-phase 3-wire installation the protectionis provided by a 250 A industrial-type circuitbreaker, the instantaneous short-circuit-current trip setting of which, is set at2,000 A (accuracy of ± 20%), i.e. in the worstcase would require 2,000 x 1,2 = 2,400 A totrip. The cable c.s.a. = 120 mm2 and theconductor material is copper.In table H1-49, the row Im = 2,000 A crossesthe column c.s.a. = 120 mm2 at the value forLmax of 296 m.The circuit breaker protects the cable againstshort-circuit faults, therefore, provided that itslength does not exceed 296 metres.

Example 2In a single-phase 230 V (phase to neutral)system, the protection is provided by a circuitbreaker with an instantaneous short-circuit-current trip setting of 500 A (± 20%), i.e.a worst case of 600 A to be certain of tripping.The cable c.s.a. = 10 mm2 and the conductormaterial is copper.In table H1-49 the row Im = 500 A crossesthe column c.s.a. = 10 mm2 at the value forLmax of 99 m.Being a 230 V single-phase circuit, acorrection factor from table H1-53 must beapplied. This factor is seen to be 0.58.The circuit breaker will therefore protect thecable against short-circuit current, providedthat its length does not exceed99 x 0.58 = 57 metres.


H1

5.2 verification of the withstand capabilities of cables under short-circuit conditions

in general verification of thethermal-withstand capability of acable is not necessary, except incases where cables of small c.s.a.are installed close to, or feedingdirectly from, the main generaldistribution board.

thermal constraintsWhen the duration of short-circuit current isbrief (several tenths of a second up to fiveseconds maximum) all of the heat producedis assumed to remain in the conductor,causing its temperature to rise. The heatingprocess is said to be adiabatic, anassumption that simplifies the calculation andgives a pessimistic result, i.e. a higherconductor temperature than that which wouldactually occur, since in practice, some heatwould leave the conductor and pass into theinsulation.For a period of 5 seconds or less, therelationship I2t = k2 S2 characterizes the timein seconds during which a conductor of c.s.a.S (in mm2) can be allowed to carry a currentI amps, before its temperature reaches alevel which would damage the surroundinginsulation.

The factor k2 is given in table H1-54 below,and is abstracted from the French standardsNF C 15-100.The method of verification consists inchecking that the thermal energy I2t per ohmof conductor material, allowed to pass by theprotecting circuit-breaker (from manufacturerscatalogues) is less than that permitted for theparticular conductor (as given in table H1-55below).

insulation conductor conductorcopper (Cu) aluminium (Al)

PVC 13225 5776PR 20449 8836table H1-54: value of the constant k 2.

S (mm2) PVC XLPEcopper aluminium copper aluminium

k 115 76 143 94k2 13225 5776 20449 88361.5 0.0297 0.0130 0.0460 0.01992.5 0.0826 0.0361 0.1278 0.05524 0.2116 0.0924 0.3272 0.14146 0.4761 0.2079 0.7362 0.318110 1.3225 0.5776 2.0450 0.883616 3.3856 1.4786 5.2350 2.262025 8.2656 3.6100 12.7806 5.522535 16.2006 7.0756 25.0500 10.824150 29.839 13.032 46.133 19.936table H1-55: maximum allowable thermal stress for cables(expressed in amperes 2 x seconds x 10 6).

Example:Is a copper-cored XLPE cable of 4 mm2 c.s.a.adequately protected by a C60N circuitbreaker (Merlin Gerin)?The above table shows that the I2t value forthe cable is 0.3272 x 106, while the maximum"let-through" value by the circuit breaker, asgiven in the manufacturer's catalogue, isconsiderably less, and amounts to0.094 x 106 ampere2-seconds.The cable is therefore adequately protectedby the circuit breaker up to its full ratedbreaking capability.

electrodynamic constraintsFor bus-trunking and other kinds ofprefabricated pre-conductored channels,rails, etc. it is necessary to verify that theelectrodynamic withstand performance whencarrying short-circuit currents is satisfactory.The peak value of current, limited by thecircuit breaker or fuse, must be less than thatfor which the pre-conductored system israted.Tables of coordination ensuring adequateprotection of their products are generallypublished by the manufacturers of suchsystems.


H1

6. protective earthing conductors (PE)

6.1 connection and choice

connection, choice and dimensioningof PE conductors (extracted fromIEC standards and the Frenchstandard NF C 15-100)

Protective (PE) conductors provide thebonding connection between all exposed andextraneous conductive parts of an installation,to create the main equipotential bondingsystem. These conductors conduct faultcurrent due to insulation failure (between aphase conductor and an exposed conductivepart) to the earthed neutral of the source. P.E.conductors are connected to the mainearthing terminal of the installation.The main earthing terminal is connected tothe earthing electrode (see Chapter F) by theearthing conductor (grounding electrodeconductor in USA).PE conductors must be:c insulated and coloured yellow and green(stripes);c be protected against mechanical andchemical damage.In IT and TN-earthed schemes it is stronglyrecommended that PE conductors should beinstalled in close proximity (i.e. in the sameconduits, on the same cable tray, etc.) as thelive cables of the related circuit. Thisarrangement ensures the minimum possibleinductive reactance in the earth-fault current-carrying circuits.

connection

PE conductors must:c not include any means of breaking thecontinuity of the circuit (such as a switch,removable links, etc.);c connect exposed conductive partsindividually to the main PE conductor, i.e.in parallel, not in series, as shown in figureH1-56;c have an individual terminal on commonearthing bars in distribution boards.TT schemeThe PE conductor need not necessarily beinstalled in close proximity to the liveconductors of the corresponding circuit, sincehigh values of earth-fault current are notneeded to operate the RCD-type of protectionused in TT installations.

IT and TN schemesThe PE or PEN conductor, as previouslynoted, must be installed as close by aspossible to the corresponding live conductorsof the circuit and no ferro-magnetic materialmust be interposed between them. A PENconductor must always be connected directlyto the earth terminal of an appliance, with alooped connection from the earth terminal tothe neutral terminal of the appliance(figure H1-57).c TN-C scheme (the neutral and PEconductor are one and the same, referred toas a PEN conductor). The protective functionof a PEN conductor has priority, so that allrules governing PE conductors apply strictlyto PEN conductors;c TN-C to TN-S transitionThe PE conductor for the installlation isconnected to the PEN terminal or bar(figure H1-58) generally at the origin of theinstallation. Downstream of the point ofseparation, no PE conductor can beconnected to the neutral conductor.

PE

PE

correct

incorrect

fig. H1-56: a poor connection in a seriesarrangement will leave all downstreamappliances unprotected.

PEN

fig. H1-57: direct connection of the PENconductor to the earth terminal of anappliance.

PEN PE

N

fig. H1-58: the TN-C-S scheme.


H1

types of materials

Materials of the kinds mentioned below intable H1-59 can be used for PE conductors,provided that the conditions mentioned in thelast column are satisfied.

type of protective earthing IT scheme TN scheme TT scheme conditions to beconductor (PE) respectedsupplementary in the same strongly strongly correct the PE conductor mustconductor cable as the recommended recommended be insulated to

phases, or in the the same levelsame cable run as the phasesindependent of the possible (1) possible (1) (2) correct c the PE conductor mayphase conductors be bare or insulated (2)

metallic housing of bus-trunking or of possible (3) PE possible (3) correctother prefabricated prewired ducting (5) PEN (8)external sheath of extruded, mineral- insulated possible (3) PE possible (3) possibleconductors (e.g. "pyrotenax" type systems) PEN not

recommended (2) (3)certain extraneous conductive elements (6) possible (4) PE possible (4) possiblesuch as: PEN forbiddenc steel building structuresc machine framesc water pipes (7)metallic cable ways, such as, conduits*, ducts, possible (4) PE possible (4) possibletrunking, trays, ladders, and so on… PEN not

recommended (2) (4)forbidden for use as PE conductors, are: metal conduits*, gas pipes, hot-water pipes, cable-armouring tapes* or wires*.

c the electrical continuitymust be assured byprotection againstdeterioration bymechanical, chemical andelectrochemical hazardsc their conductancemust be adequate

* forbidden in some countries only-universally allowed to be used for supplementary equipotential conductors.

table H1-59: choice of protective conductors (PE).

(1) In schemes TN and IT, fault clearance isgenerally effected by overcurrent devices(fuses or circuit breakers) so that theimpedance of the fault-current loop must besufficiently low to assure positive protectivedevice operation. The surest means ofachieving a low loop impedance is to use asupplementary core in the same cable as thecircuit conductors (or taking the same routeas the circuit conductors). This stratagemminimizes the inductive reactance andtherefore the impedance of the loop.(2) The PEN conductor is a neutral conductorthat is also used as a protective earthconductor. This means that a current may beflowing through it at any time (in the absenceof an earth fault). For this reason an insulatedconductor is recommended for PENoperation.(3) The manufacturer provides the necessaryvalues of R and X components of theimpedances (phase/PE, phase/PEN) toinclude in the calculation of the earth-fault

loop impedance.(4) Possible, but not recomended, since theimpedance of the earth-fault loop cannot beknown at the design stage. Measurements onthe completed installation are the onlypractical means of assuring adequateprotection for persons.(5) It must allow the connection of other PEconductors.NB: these elements must carry an indivualgreen/yellow striped visual indication, 15 to100 mm long (or the letters PE at less than15 cm from each extremity).(6) These elements must be demountableonly if other means have been provided toensure uninterrupted continuity of protection.(7) With the agreement of the appropriatewater authorities.(8) In the prefabricated pre-wired trunkingand similar elements, the metallic housingmay be used as a PEN conductor, in parallelwith the corresponding bar, or other PEconductor in the housing.

6.2 conductor dimensioning

Table H1-60 below is based on the Frenchnational standard NF C 15-100 for LVinstallations. This table provides two methodsof determining the appropriate c.s.a. for bothPE or PEN conductors, and also for theconductor to the earth electrode.The two methods are:c adiabatic (which corresponds with thatdescribed in IEC 724)This method, while being economical andassuring protection of the conductor againstoverheating, leads to small c.s.a.'s comparedto those of the corresponding-circuit phaseconductors. The result is sometimesincompatible with the necessity in IT and TNschemes to minimize the impedance of thecircuit earth-fault loop, to ensure positiveoperation by instantaneous overcurrenttripping devices. This method is used in

practice, therefore, for TT installations, andfor dimensioning an earthing conductor*.c simplifiedThis method is based on PE conductor sizesbeing related to those of the corresponding-circuit phase conductors, assuming that thesame conductor material is used in eachcase.Thus, in table H1-60 for:Sph i 16 mm2 SPE = Sph16 < Sph i 35 mm2 SPE = 16 mm2

Sph > 35 mm2 SPE = Sph/2c note: when, in a TT scheme , theinstallation earth electrode is beyond thezone of influence of the source earthingelectrode, the c.s.a. of the PE conductor canbe limited to 25 mm2 (for copper) or 35 mm2

(for aluminium).* grounding electrode conductor.


H1

k values nature of insulationpolyvinylchloride (PVC) cross-linked-polyethylene

(XLPE)ethylene-propylene-rubber(EPR)

final 160 250temperature ( °C)insulated conductors initial temperature initial temperaturenot incoporated in cables θ initial = 30 °C θ initial = 30 °Cor bare conductorsin contact withcable jackets

copper 143 176aluminium 95 116steel 52 64

conductors of a initial temperature initial temperaturemulti-core-cable θ initial = 30 °C θ initial = 30 °C

copper 115 143aluminium 76 94

6. protective earthing conductors (PE) (continued)

The neutral cannot be used as a PENconductor unless its c.s.a. is equal to orlarger than 10 mm2 (copper) or 16 mm2

(aluminium).Moreover, a PEN conductor is not allowed ina flexible cable. Since a PEN conductorfunctions also as a neutral conductor, itsc.s.a. cannot, in any case, be less than thatnecessary for the neutral, as discuss in Sub-clause 7.1 of this Chapter.This c.s.a. cannot be less than that of thephase conductors unless:

6.2 conductor dimensioning (continued)

c the kVA rating of single-phase loads is lessthan 10% of the total kVA load, andc Imax likely to pass through the neutral innormal circumstances, is less than thecurrent permitted for the cable size selected.Furthermore, protection of the neutralconductor must be assured by the protectivedevices provided for phase-conductorprotection (described in Sub-clause 7.2of this Chapter).

c.s.a. of phase c.s.a. of c.s.a. of c.s.a. of earthing conductor betweenconductors Sph (mm 2) PE conductor PEN conductor the installation earth electrode

Cu Alu and the main earth terminalsimplified i 16 i 16 SPE = Sph (1) c when protected against mechanicalmethod 25 SPE = 16 damage: S = I √t (2)

25,35 35 k> 35 > 35 SPE = Sph/2

adiabatic any size SPE = I √t (1) (2)method k

SPEN = Sph with minimum10/mm2 Cu, 16/mm2 AluSPEN = Sph/2 à Sph (3) withminimum 16/mm2 Cu, 25/mm2 Alu c without mechanical protection, but

protected against corrosion byimpermeable cable sheath. Minimum16 mm2 for copper or galvanized steel.c without either of the above protections;minimum of 25 mm2 for bare copper and50 mm2 for bare galvanized steel.

table H1-60: minimum c.s.a.'s for PE conductors and earthing conductors (to the installation earth electrode).

(1) When the PE conductor is separated from the circuit phase conductors, the followingminimum values must be respected:c 2.5 mm2 if the PE is mechanically protectedc 4 mm2 if the PE is not mechanically protected.(2) Refer to table H1-55 for the application of this formula.(3) According to the conditions prescribed in the introduction to this table.

values of factor k to be used in

the formulae (2)

These values are identical in several nationalstandards, and the temperature rise ranges,together with factor k values and the uppertemperature limits for the different classesof insulation, correspond with those publishedin IEC 724 (1984).

The data presented in table H1-61 are thosemost commonly needed for LV installationdesign.

table H1-61: k factor values for LV PE conductors, commonly used in national standardsand complying with IEC 724.


H1

6.3 protective conductor between the HV/LV transformer and the main general distributionboard (MGDB)

All phase and neutral conductors upstream ofthe main incoming circuit breaker controllingand protecting the MGDB are protected bydevices at the HV side of the transformer.The conductors in question, together with thePE conductor, must be dimensionedaccordingly. Dimensioning of the phase andneutral conductors from the transformer isexemplified in Sub-clause 1.6 of this chapter(for circuit C1 of the system illustrated infig. H1-8).Recommended conductor sizes for bare andinsulated PE conductors from the transformerneutral point, shown in figure H1-62, areindicated below in table H1-63. The kVArating to consider is the sum of all (if morethan one) transformers connected to theMGDB.The table indicates the c.s.a. of theconductors in mm2 according to:c the nominal rating of the HV/LVtransformer(s) in kVA;c the fault-current clearance time by the HVprotective devices, in seconds,

PEMGDB

main earth barfor the LV installation

fig. H1-62: PE conductor to the main earthbar in the MGDB.

c the kinds of insulation and conductormaterials.If the HV protection is by fuses, then use the0.2 seconds columns.In IT schemes, if an overvoltage protectiondevice is installed (between the transformerneutral point and earth) the conductors forconnection of the device should also bedimensioned in the same way as thatdescribed above for PE conductors.

P conductor conductors conductors conductors(kVA) material bare PVC-insulated XLPE-insulatedLV voltages copper t(s) 0.2 s 0.5 s - 0.2 s 0.5 s - 0.2 s 0.5 s -127/ 230/ aluminium t(s) - 0.2 s 0.5 s - 0.2 s 0.5 s - 0.2 s 0.5 s220 V 400 Vi 63 i 100 c.s.a. of PE 25 25 25 25 25 25 25 25 25100 160 conductors 25 25 35 25 25 50 25 25 35125 200 SPE (mm2) 25 35 50 25 35 50 25 25 50160 250 25 35 70 35 50 70 25 35 50200 315 35 50 70 35 50 95 35 50 70250 400 50 70 95 50 70 95 35 50 95315 500 50 70 120 70 95 120 50 70 95400 630 70 95 150 70 95 150 70 95 120500 800 70 120 150 95 120 185 70 95 150630 1000 95 120 185 95 120 185 70 120 150800 1250 95 150 185 120 150 240 95 120 185

these conductors must bedimensioned according to nationalpractices.

table H1-63: c.s.a. of PE conductor between the HV/LV transformer and the MGDB, interms of transformer ratings and fault-clearance times used in France.

6.4 equipotential conductorThe main equipotential conductorThis conductor must, in general, have a c.s.a.at least equal to a half of that of the largestPE conductor, but in no case need exceed25 mm2 (copper) or 35 mm2 (aluminium)while its minimum c.s.a. is 6 mm2 (copper) or10 mm2 (aluminium).

Supplementary equipotential conductorThis conductor allows an exposed conductivepart which is remote from the nearest mainequipotential conductor (PE conductor) to beconnected to a local protective conductor. Itsc.s.a. must be at least a half of that of theprotective conductor to which it is connected.If it connects two exposed conductive parts(M1 and M2 in figure H1-64) its c.s.a. must beat least equal to that of the smaller of the twoPE conductors (for M1 and M2). Equipotentialconductors which are not incorporated in acable, should be protected mechanically byconduits, ducting, etc. wherever possible.Other important uses for supplementaryequipotential conductors concern thereduction of the earth-fault-loop impedance,particulary for indirect-contact protectionschemes in TN- or IT-earthed installations,and in special locations with increasedelectrical risk (IEC 364-4-41 refers).

– between two exposed conductive parts,if SPE1 i SPE2if SLS = SPE1

– between an exposed conductive part and a metallic structure

(*) with a minimum of 2.5 mm2 for mechanically protected conductors ;4 mm2 for conductors not mechanically protected copper equivalent

M1 M2

SPE1 SPE2SLS

SLS =SPE

2SPE1

SLS

metal structures(conduits, girders, …)

M1

fig. H1-64: supplementary equipotentialconductors.


H1

7. the neutral conductor

The c.s.a. and the protection of the neutralconductor, apart from its current-carryingrequirement, depend on several factors,namely:c the type of earthing system, TT, TN, etc.;c method of protection against indirect-contact hazards according to the methodsdescribed below.

7.1 dimensioning the neutral conductor

influence of the type of earthingsystemTT, TN-S and IT schemesc single-phase circuits or those of c.s.a.i 16 mm2 (copper) 25 mm2 (aluminium): thec.s.a. of the neutral conductor must be equalto that of the phases;c three-phase circuits of c.s.a. > 16 mm2

copper or 25 mm2 aluminium:the c.s.a. of the neutral may be chosen to be:v equal to that of the phase conductors, orv smaller, on condition that:- the current likely to flow through the neutralin normal conditions is less than the permittedvalue Iz. The influence of triplen* harmonicsmust be given particular consideration, aspreviously noted in Chapter F Sub-clause2.2, or:- the single-phase power of the circuit is lessthan 10% of the balanced 3-phase power ofthe circuit, or:- the neutral conductor is protected againstshort-circuit, in accordance with the followingSub-clause H1-7.2.* the 3rd and multiples of the 3rd harmonic.

TN-C schemeThe same conditions apply in theory as thosementioned above, but in practice, the neutralconductor must not be open-circuited underany circumstances since it constitutes a PEas well as a neutral conductor (see table H1-60 "c.s.a. of PEN conductor" column).IT schemeIn general, it is not recommended to distributethe neutral conductor, i.e. a 3-phase 3-wirescheme is preferred.

When a 3-phase 4-wire installation isnecesssary, however, the conditionsdescribed above for TT and TN-S schemesare applicable.

7.2 protection of the neutral conductor

Table H1-65 summarizes the several possiblecases. The table has been based on Frenchnational standards (NF C 15-100). Thefollowing points however, should beconsidered when referring to the table.Circuit isolationIt is considered to be good practice that everycircuit be provided with the means for itsisolation.

Circuit breakingTable H1-65 is based on circuit breakers,which in the event of a fault, will open allpoles, including the neutral pole, i.e. thecircuit breakers are omnipolar. This table canalso be used for fuses able to emulate thisomnipolar opening.The action can only be achieved with fuses inan indirect way, in which the operation of oneor more fuses provokes a mechanical trip-outof all poles of an associated series-connectedload-break switch.The action is commonly caused by a striker-pin which is projected by means of anexplosive cartridge (triggered by the blowingfuse) against the switch tripping mechanism.Reclosing of the switch however, must bepossible only when the used cartridge hasbeen replaced by a new one.

Protection against electric shocksTable H1-65 takes into account the fact thatprotection against indirect-contact dangersdepend either on 300 mA RCDs (TT system)or on circuit breakers (TN and IT systems).


H1

earthing systemsTT TN-C TN-S IT

reminder: protection by RCD provided by circuit according to the provided by circuit breakersagainst indirect breakers or fuses with method of protection or fuses and one RCD at leastcontact Ia (fuses) or Im (CB) chosen for each group of appliances

< Isc (min) connected to an earthelectrode (see figure G20)

protected circuit1-phase P-Nphase/neutral

1-phase 2Pphase/phase

3-phase 3P3-wire

3-phase 3P-N4-wire Sn = Sph*

3P-NSn < Sph

(A)

(B)

(A)

(B)

(C)

(C)

(C)

* Sn = c.s.a. of neutral conductorSph = c.s.a. of phase

table H1-65: table of protection schemes for neutral conductors in different earthing systems.

Symbol for overcurrent and short-circuittripping devices.(A) authorized for TT and TN schemes if aRCD is installed at the origin of the circuit orupstream of it, and if no artificial neutral isdistributed downstream of its location.(B) authorized for TT and TN schemes if theneutral conductor is protected against short-circuits by protective arrangements made forthe phases, and if the normal service currentis substantially less than the maximumpermissible for the neutral conductorconcerned.(C) authorized for IT schemes in certainconditions, viz: if the circuit breakercontrolling a number of homogeneous (i.e.similar) final circuits, of which the ratio of theextreme circuit ratings does not exceed 2,and which are protected against a secondfault occurring elsewhere in the installation bya RCD of sensitivity i 15% of that of thecalibration of the final circuit having thesmallest c.s.a.Refer to example 2 for CB 5.

thermal magnetic


H1

7. the neutral conductor

7.2 protection of the neutral conductor (continued)

examplesExample 1: (figure H1-66)3-phase 4-wire circuit with 3 x 95 mm2 copperphase conductors and 1 x 50 mm2 copperneutral conductor. The installation isTT-earthed with RCD protection upstream.Single-phase power of load: 70 kVA(connected phase-neutral). Three-phasepower of load: 140 kVA. The condition ofsingle-phase power being < 10% of the3-phase power delivered (Sub-clause 7.1 TTand TN-S schemes) is not satisfied in thiscase, since 70/140 = 50%. A reduced neutralc.s.a. may however, be used, provided that itis correctly protected. A suitable circuitbreaker for this purpose would be a 4-poleunit rated at 250 A with 3 trip units (1 for eachphase) set at 250 A and 1 trip unit (for theneutral) set at 125 A. The operation of anyone (or more) of these tripping units will tripall four poles of the circuit breaker.

N

4-pole CB3 - 250 A1 - 125 A trip units

50 mm2

3-phase power140 kVA

1-phase power70 kVA

3-phaseloads

3 x 95 mm2

1-phaseloads


Example 2: (fig. H1-67)An installation is IT-earthed with a distributedneutral (i.e. a 3-phase 4-wire system in whichthe neutral is not earthed). This arrangementis not recommended, particularly for small ormedium-sized installations, but it doesprovide two levels of voltage, e.g. 230 V and400 V. The interposition of a LV/LVtransformer in a 3-phase 3-wire IT scheme(as shown in figure H1-8) is a preferredmethod of obtaining the two levels of voltage,and in such a case, the neutral conductormay be earthed.

Circuit breaker 1, 2 and 3As in example 1, the circuits protected bythese CBs have a neutral conductor of 50%of the c.s.a. of the phase conductors. Thecircuit breakers will therefore be 4-pole units,with tripping devices similar to thosementioned in example 1.

Circuit breaker 5This arrangement corresponds with thatdescribed in (C) of table H1-65 concerning acircuit breaker connected directly to, andcontrolling the busbars from which a numberof similar final circuits, (protected by 2-pole,1-phase and neutral CBs) are supplied.Overcurrent tripping is provided on all out-going CBs, but the 4-pole incoming CB (no 5)has only the (300 mA) RCD protection(mentioned in (C) of table H1-65) themagnetic core of which embraces all4 conductors.It may be noted that circuit breaker 12supplies a long lighting circuit, the neutral andphase conductors of which have the samec.s.a. A 4-pole circuit breaker having 3tripping devices is therefore suitable (1 deviceper phase).

400 kVAHV LV

PIMcompact NS250N4-pole3d 250 A1d 125 A

3 x 120 mm2+1 x 70 mm2

3 x 185 mm2+1 x 95 mm2

NS100N3-pole3d 100 A

3 x 35 mm2

4 x 6 mm2 4 x 50 mm2 4 x 6 mm2 contactorLC1 D63

thermaloverloadrelay

3 x 16 mm2

30 kW58 A

3 x 50 mm2

PE PE

PE

2 single-core cables 120 mm2 per phase - 1 x 120 mm2 (neutral)

PEPE44 5 7 8 9 1011

12

4 x 70 mm2

2 x

2.5

mm

2

2 x

1.5

mm

2

2 x

1.5

mm

2

4 x

2.5

mm

2

DPN

PEN

6

PE

DPN DPN DPN DPN

NS100N4-pole 32 A300 mA

C60N4-pole4d 32 Adiff. HS300 mA


C60N4-pole4d 32 A


NS80HMA NS160N4-pole + MT100/1603d 160 Adiff. BS10 A

compact NS400N4-pole3d 250 A1d 125 A

outdoorlighting

3d = 3 tripping units4d = 4 tripping unitsdiff. HS = high-sensitivity differential trippingdiff. BS = low-sensitivity differential tripping



H1

x In

t

4In

fuse-blowncurve

minimumpre-arcingtime curve

Example 3: (figure H1-68)TN-C/TN-S installationThree-pole circuit breakers only must beused for nos. 1, 2, 3 and 7, since noswitchgear of any kind must be included inthe combined protective and neutralconductor (PEN) associated with them.The total single-phase power of the load isless than 10% of the 3-phase power, so thatthe c.s.a. of the PEN conductor from thesource (i.e. circuit 1) may be half that of thephase conductors of the circuit. Theprotection against indirect contact for thiscircuit (1) is provided by CB1 if the maximumlength of the circuit is less than Lmax (seeChapter G Sub-clause 5.2).For a 630 A CB, regulated to tripinstantaneously at a current level of 4 In

L max = 0.8 x 230 x 240 x 103 22.5 (1.25 + 2) x IaThe 1.25 factor in the denominator is a 25%increase in resistance for a 240 mm2 c.s.a.conductor*, while Ia = 630 x 4 x 1.15 wherethe factor 1.15 allows for the guaranteed± 15% tolerance of the instantaneousmagnetic tripping element of the circuitbreaker (i.e. the worst case, requiring theshortest Lmax).Lmax = 208 metres.* Chapter G Sub-clause 5.2

Section of the installation which is TN-Sconnected (PE conductor and neutralconductor separated at a point upstream)Circuit breaker 4c.s.a. neutral = c.s.a. phase, so that a trippingdevice for the neutral current is notnecessary. A 4-pole circuit breaker with onetripping device per phase is thereforeappropriate.Circuit breaker 5c.s.a. neutral = 50% c.s.a. phase, so that atripping device for the neutral is required.A 4-pole circuit breaker with 3 trippingdevices (set at 160 A) for the phases, and1 tripping device for the neutral (set at 80 A)is required, as noted in (B) of table H1-65.Circuit breaker 6The protection of a circuit supplying socketoutlets, as mentioned frequently in earlierChapters, must include a RCD of highsensitivity (generally of 30 mA).Associated circuit breaker and contactor 8This combination provides short-circuitprotection (circuit breaker) and overloadprotection (thermal relays on contactor to suitmotor characteristics). The circuit breaker hasno thermal tripping devices, while thecontactor has three (one for each phase).Circuit breaker 9Controls and protects an extensive lightingcircuit, and since the phase and neutralconductors have the same c.s.a. a 4-polecircuit breaker is suitable, having 3 trippingdevices (1 for each phase).



H2

1. the basic functions of LV switchgear

the role of switchgear is that of:c electrical protection;c safe isolation from live parts;c local or remote switching.

National and international standards definethe manner in which electric circuits of LVinstallations must be realized, and thecapabilities and limitations of the variousswitching devices which are collectivelyreferred to as switchgear.The main functions of switchgear are:c electrical protection;c electrical isolation of sections of aninstallation;c local or remote switching.These functions are summarized below intable H2-1.Electrical protection at low voltage is (apartfrom fuses) normally incorporated in circuit

breakers, in the form of thermal-magneticdevices and/or residual-current-operatedtripping devices (less-commonly, residual-voltage-operated devices - acceptable to, butnot recommended by IEC).In addition to those functions shown in tableH2-1, other functions, namely:c over-voltage protection;c under-voltage protection are provided byspecific devices (lightning and various othertypes of voltage-surge arrester; relaysassociated with: contactors, remotely-controlled circuit breakers, and with combinedcircuit breaker/isolators… and so on).

electrical protection isolation controlagainstoverload currents - isolation clearly indicated - functional switchingshort-circuit currents by an authorized fail-proof - emergency switchinginsulation failure mechanical indicator - emergency stopping

- a gap or interposed insulating - switching off forbarrier between the open mechanical maintenancecontacts, clearly visible.

table H2-1: basic functions of LV switchgear.

1.1 electrical protection

electrical protection assures:c protection of circuit elementsagainst the thermal and mechanicalstresses of short-circuit currents;c protection of persons in the eventof insulation failure;c protection of appliances andapparatus being supplied (e.gmotors, etc.).

The aim is to avoid or to limit the destructiveor dangerous consequences of excessive(short-circuit) currents, or those due tooverloading and insulation failure, and toseparate the defective circuit from the rest ofthe installation.A distinction is made between the protectionof:c the elements of the installation (cables,wires, switchgear…);c persons and animals;c equipment and appliances supplied fromthe installation;c the protection of circuits(see chapter H1):v against overload; a condition of excessivecurrent being drawn from a healthy(unfaulted) installation,v against short-circuit currents due tocomplete failure of insulation betweenconductors of different phases or (in TNsystems) between a phase and neutral (orPE) conductor.Protection in these cases is provided eitherby fuses or circuit breaker, at the distributionboard from which the final circuit (i.e. the

circuit to which the load is connected)originates. Certain derogations to this rule areauthorized in some national standards, asnoted in chapter H1 sub-clause 1.4.c the protection of persons againstinsulation failures (see chapter G).According to the system of earthing for theinstallation (TN, TT or IT) the protection willbe provided by fuses or circuit breakers,residual current devices, and/or permanentmonitoring of the insulation resistance of theinstallation to earth.c the protection of electric motors(see chapter J clause 5) against overheating,due, for example, to long term overloading;stalled rotor; single-phasing, etc. Thermalrelays, specially designed to match theparticular characteristics of motors are used.Such relays may, if required, also protect themotor-circuit cable against overload. Short-circuit protection is provided either by typeaM fuses or by a circuit breaker from whichthe thermal (overload) protective element hasbeen removed, or otherwise madeinoperative.

1.2 isolation

a state of isolation clearly indicatedby an approved "fail-proof" indicator,or the visible separation of contacts,are both deemed to satisfy thenational standards of manycountries.

The aim of isolation is to separate a circuit orapparatus, or an item of plant (such as amotor, etc.) from the remainder of a systemwhich is energized, in order that personnelmay carry out work on the isolated part inperfect safety.In principle, all circuits of an LV installationshall have means to be isolated. In practice,in order to maintain an optimum continuity ofservice, it is preferred to provide a means ofisolation at the origin of each circuit.An isolating device must fulfil the followingrequirements:c all poles of a circuit, including the neutral(except where the neutral is a PENconductor) must be open (1);

c it must be provided with a means of lockingopen with a key (e.g. by means of a padlock)in order to avoid an unauthorized reclosureby inadvertence;c it must conform to a recognized national orinternational standard (e.g. IEC 947-3)concerning clearance between contacts,creepage distances, overvoltage withstandcapability, etc. and also:(1) the concurrent opening of all live conductors, while notalways obligatory, is however, strongly recommended (forreasons of greater safety and facility of operation). Theneutral contact opens after the phase contacts, and closesbefore them (IEC 947-1).


H2

1. the basic functions of LV switchgear (continued)

1.2 isolation (continued)

v verification that the contacts of the isolatingdevice are, in fact, open. The verification maybe:- either visual, where the device is suitablydesigned to allow the contacts to be seen(some national standards impose thiscondition for an isolating device located at theorigin of a LV installation supplied directlyfrom a HV/LV transformer),- or mechanical, by means of an indicatorsolidly welded to the operating shaft of thedevice. In this case the construction of thedevice must be such that, in the eventualitythat the contacts become welded together inthe closed position, the indicator cannotpossibly indicate that it is in the open position.v leakage currents. With the isolating deviceopen, leakage currents between the opencontacts of each phase must not exceed:- 0.5 mA for a new device,- 6.0 mA at the end of its useful life.v voltage-surge withstand capability, acrossopen contacts. The isolating device, whenopen must withstand a 1.2/50 µs impulse,having a peak value of 5, 8 or 10 kVaccording to its service voltage, as shown intable H2-2. The device must satisfy theseconditions for altitudes up to 2,000 metres.Consequently, if tests are carried out at sealevel, the test values must be increased by23% to take into account the effect of altitude.See standard IEC 947 and the Noteimmediately preceding table F-10.

service (nominal) impulse withstandvoltage peak voltage(V) (kV)230/400 5 kV400/690 8 kV1,000 10 kV

table H2-2: peak value of impulse voltageaccording to normal service voltage oftest specimen.

Industrial LV switchgear which affordsisolation when open is marked on the frontface by the symbol .This symbol may be combined with thoseindicating other features where a device alsoperforms other functions as shown in figureH2-4.

fig. H2-3: symbol for a disconnector* alsocommonly referred to as an isolator.

switch-disconnector*, also referred to as a load-break isolating switch

circuit breaker suitable for circuit isolation

fig. H2-4: symbols for circuit isolationcapability incorporated in other switchingdevices.

* IEC 617-7 and 947-3.

Note. In this guide the terms "disconnector"and "isolator" have the same meaning.

1.3 switchgear control

switchgear-control functions allowsystem operating personnel tomodify a loaded system at anymoment, according to requirements,and include:c functional control (routineswitching, etc.);c emergency switching;c maintenance operations on thepower system.

In broad terms "control" signifies any facilityfor safely modifying a load-carrying powersystem at all levels of an installation. The

operation of switchgear is an important part ofpower-system control.

functional control

This control relates to all switching operationsin normal service conditions for energizing orde-energizing a part of a system orinstallation, or an individual piece ofequipment, item of plant, etc.Switchgear intended for such duty must beinstalled at least:c at the origin of any installation;c at the final load circuit or circuits (oneswitch may control several loads).Marking (of the circuits being controlled) mustbe clear and unambiguous.In order to provide the maximum flexibilityand continuity of operation, particularly wherethe switching device also constitutes theprotection (e.g. a circuit breaker orswitch-fuse) it is preferable to include aswitch at each level of distribution, i.e. on

each outgoing way of all distribution and sub-distribution boards.

The manœuvre may be:c either manual (by means of an operatinglever on the switch) or;c electric, by push-button on the switch or ata remote location (load-shedding and re-connection, for example).These switches operate instantaneously (i.e.with no deliberate delay), and those thatprovide protection are invariably omni-polar*.The main circuit breaker for the entireinstallation, as well as any circuit breakersused for change-over (from one source toanother) must be omni-polar units.* one break in each phase and (where appropriate) onebreak in the neutral (see table H1-65).


H2

emergency switching -

emergency stop

An emergency switching is intended tode-energize a live circuit which is, or couldbecome, dangerous (electric shock or fire).An emergency stop is intended to arrest amovement which has become dangerous. Inthe two cases:c the emergency control device or its meansof operation (local or at remote location(s))such as a large red mushroom-headedemergency-stop pushbutton must berecognizable and readily accessible, inproximity to any position at which dangercould arise or be seen;c a single action must result in a completeswitching-off of all live conductors (1) (2);

c a "break glass" emergency switchinginitiation device is authorized, but inunmanned installations the re-energizing ofthe circuit can only be achieved by means ofa key held by an authorized person.It should be noted that in certain cases, anemergency system of braking, may requirethat the auxiliary supply to the braking-systemcircuits be maintained until final stoppage ofthe machinery.(1) Taking into account stalled motors.(2) In a TN schema the PEN conductor must never beopened, since it functions as a protective earthing wire aswell as the system neutral conductor.

switching-off for mechanical

maintenance work

This operation assures the stopping of amachine and its impossibility to beinadvertently restarted while mechanicalmaintenance work is being carried out on thedriven machinery. The shutdown is generallycarried out at the functional switching device,with the use of a suitable safety lock andwarning notice at the switch mechanism.


H2

2. the switchgear and fusegear

2.1 elementary switching devices

disconnector (or isolator)

This switch is a manually-operated, lockable,two-position device (open/closed) whichprovides safe isolation of a circuit whenlocked in the open position. Its characteristicsare defined in IEC 947-3.A disconnector is not designed to make or tobreak current* and no rated values for thesefunctions are given in standards. It must,however, be capable of withstanding thepassage of short-circuit currents and isassigned a rated short-time withstandcapability; generally for 1 second, unlessotherwise agreed between user andmanufacturer. This capability is normallymore than adequate for longer periods of(lower-valued) operational overcurrents, suchas those of motor-starting.Standardized mechanical-endurance,overvoltage, and leakage-current tests, mustalso be satisfied.* i.e. a LV disconnector is essentially a dead-system switching device to be operated withno voltage on either side of it, particularlywhen closing, because of the possibility of anunsuspected short-circuit on the downstreamside. Interlocking with an upstream switch orcircuit breaker is frequently used.

fig. H2-5: symbol for a disconnector(or isolator).

load-breaking switch

This control switch is generally operatedmanually (but is sometimes provided withelectrical tripping for operator convenience)and is a non-automatic two-position device(open/closed).It is used to close and open loaded circuitsunder normal unfaulted circuit conditions.It does not consequently, provide anyprotection for the circuit it controls.IEC standard 947-3 defines:c the frequency of switch operation(600 close/open cycles per hour maximum);c mechanical and electrical endurance(generally less than that of a contactor);c current making and breaking ratings fornormal and infrequent situations.IEC 947-3 also recognizes 3 categories ofload-breaking switch, each of which issuitable for a different range of load powerfactors, as shown in table H2-7.

fig. H2-6: symbol for a load-breakingswitch.

When closing a switch to energize a circuitthere is always the possibility that an(unsuspected) short circuit exists on thecircuit. For this reason, load-break switchesare assigned a fault-current making rating,i.e. successful closure against theelectrodynamic forces of short-circuit currentis assured. Such switches are commonlyreferred to as "fault-make load-break"switches.Upstream protective devices are relied uponto clear the short-circuit fault.


H2

nature utilization category typical applicationsof frequent infrequentcurrent operation operationalternating AC-20A AC-20B connecting and disconnectingcurrent under no-load conditions

AC-21A AC-21B switching of resistive loadsincluding moderate overloads

AC-22A AC-22B switching of mixed resistiveand inductive loads, includingmoderate overloads

AC-23A AC-23B switching of motor loads orother highly inductive loads

table H2-7: utilization categories of LV a.c. switches according to IEC 947-3.

Category AC-23 includes occasionalswitching of individual motors. The switchingof capacitors or of tungsten filament lampsshall be subject to agreement betweenmanufacturer and user.The utilization categories referred to in tableH2-7 do not apply to an equipment normallyused to start, accelerate and/or stopindividual motors. The utilization categoriesfor such an equipment are dealt with inchapter J, table J5-4.

Example:A 100 A load-break switch of category AC-23(inductive load) must be able:c to make a current of 10 In (= 1,000 A) at apower factor of 0.35 lagging;c to break a current of 8 In (= 800 A) at apower factor of 0.35 lagging;c to withstand short-circuit currents (not lessthan 12 In) passing through it for 1 second,where 12 In equals the r.m.s. value of the a.c.component, while the peak value (expressedin kA) is given by a factor "n" in table XVI ofIEC 947- Part 1, reproduced below for readerconvenience (table H2-8).

test current I power-factor time-constant n(A) (ms)

I i 01 500 0.95 5 1.41 1 500 < I i 3 000 0.9 5 1.42 3 000 < I i 4 500 0.8 5 1.47 4 500 < I i 6 000 0.7 5 1.53 6 000 < I i 10 000 0.5 5 1.710 000 < I i 20 000 0.3 10 2.020 000 < I i 50 000 0.25 15 2.150 000 < I 0.2 15 2.2

table H2-8: factor "n" used for peak-to-rms value (IEC 947-part1).

bistable switch (télérupteur)

This device is extensively used in the controlof lighting circuits where the depression of apushbutton (at a remote control position) willopen an already-closed switch or close anopen switch in a bistable sequence.Typical applications are:c two-way switching on stairways of largebuildings;c stage-lighting schemes;c factory illumination, etc.Auxiliary devices are available to provide:c remote indication of its state at any instant;c time-delay functions;c maintained-contact features.

fig. H2-9: symbol for a bistable remotely-operated switch (télérupteur) .


H2

2. the switchgear and fusegear (continued)

contactor

The contactor is a solenoid-operatedswitching device which is generally heldclosed by (a reduced) current through theclosing solenoid (although variousmechanically-latched types exist for specificduties). Contactors are designed to carry outnumerous close/open cycles and arecommonly controlled remotely by on-offpushbuttons.The large number of repetitive operatingcycles is standardized in table VIII of IEC947-4-1 by:c the operating duration: 8 hours;uninterrupted; intermittent; temporary of 3,10, 30, 60 and 90 minutes;c utilization category: (for definition see tableJ5-4) for example, a contactor of categoryAC3 can be used for the starting andstopping of a cage motor;c the start-stop cycles (1 to 1,200 cyles perhour);c mechanical endurance (number of off-loadmanœuvres);c electrical endurance (number of on-loadmanœuvres);

c a rated current making and breakingperformance according to the category ofutilization concerned.

Example:A 150 A contactor of category AC3 must havea minimum current-breaking capability of 8 In(= 1,200 A) and a minimum current-makingrating of 10 In (= 1,500 A) at a power factor(lagging) of 0.35.

controlcircuit

power circuit

fig. H2-10: symbol for a contactor.

discontactor*

A contactor equipped with a thermal-typerelay for protection against overloadingdefines a "discontactor". Discontactors areused extensively for remote push-buttoncontrol of lighting circuits, etc., and may alsobe considered as an essential element in amotor controller, as noted in sub-clause 2.2."combined switchgear elements".The discontactor is not the equivalent of a

circuit breaker, since its short-circuit current-breaking capability is limited to 8 or 10 In. Forshort-circuit protection therefore, it isnecessary to include either fuses or a circuitbreaker in series with, and upstream of, thediscontactor contacts.*This term is not defined in IEC publicationsbut is commonly used in some countries.

fuses

Fuses exist with and without "fuse-blown"mechanical indicators.Fuses break a circuit by controlled melting ofthe fuse element when a current exceeds agiven value for a corresponding period oftime; the current/time relationship beingpresented in the form of a performance curvefor each type of fuse.Standards define two classes of fuse:c those intended for domestic installations,manufactured in the form of a cartridge forrated currents up to 100 A and designatedtype gG in IEC 269-3;c those for industrial use, with cartridge typesdesignated gG (general use); and gM and aM(for motor-circuits) in IEC 269-1 and 2.The main differences between domestic andindustrial fuses are the nominal voltage andcurrent levels (which require much largerphysical dimensions) and their fault-currentbreaking capabilities.Type gG fuse-links are often used for theprotection of motor circuits, which is possiblewhen their characteristics are capable ofwithstanding the motor-starting currentwithout deterioration.A more recent development has been theadoption by the IEC of a fuse-type gM formotor protection, designed to cover starting,and short-circuit conditions. This type of fuseis more popular in some countries than inothers, but at the present time the aM fuse incombination with a thermal overload relay ismore-widely used.

two classes of LV cartridge fuse arevery widely used:c for domestic and similarinstallations type gGc for industrial installations type gG,gM or aM.

A gM fuse-link, which has a dual rating ischaracterized by two current values. The firstvalue In denotes both the rated current of thefuse-link and the rated current of the fuse-holder; the second value Ich denotes thetime-current characteristic of the fuse-link asdefined by the gates in Tables II, III and VI ofIEC 269-1.These two ratings are separated by a letterwhich defines the applications.For example: In M Ich denotes a fuseintended to be used for protection of motorcircuits and having the characteristic G. Thefirst value In corresponds to the maximumcontinuous current for the whole fuse and thesecond value Ich corresponds to the Gcharacteristic of the fuse link. For furtherdetails see note at the end of sub-clause 2.1.An aM fuse-link is characterized by onecurrent value In and time-currentcharacteristic as shown in figure H2-14.Important: Some national standards use a gI(industrial) type fuse, similar in all mainessentails to type gG fuses.Type gI fuses should never be used,however, in domestic and similar installations.

fig. H2-11: symbol for fuses.

2.1 elementary switching devices (continued)


H2

fusing zones -

conventional currents

The conditions of fusing (melting) of a fuseare defined by standards, according to theirclass.c class gG fusesThese fuses provide protection againstoverloads and short-circuits.Conventional non-fusing and fusing currentsare standardized, as shown in figure H2-12and in table H2-13.v the conventional non-fusing current Inf isthe value of current that the fusible elementcan carry for a specified time without melting.Example: a 32 A fuse carrying a current of1.25 In (i.e. 40 A) must not melt in less thanone hour (table H2-13)v the conventional fusing current If (=I2 infig.H2-12) is the value of current which willcause melting of the fusible element beforethe expiration of the specified time.Example: a 32 A fuse carrying a current of1.6 In (i.e. 52.1 A) must melt in one hour orless (table H2-13).IEC 269-1 standardized tests require that afuse-operating characteristic lies between thetwo limiting curves (shown in figure H2-12) forthe particular fuse under test. This meansthat two fuses which satisfy the test can havesignificantly different operating times at lowlevels of overloading.v the two examples given above for a 32 Afuse, together with the foregoing notes onstandard test requirements, explain why

these fuses have a poor performance in thelow overload range.v it is therefore necessary to install a cablelarger in ampacity than that normally requiredfor a circuit, in order to avoid theconsequences of possible long termoverloading (60% overload for up to one hourin the worst case).By way of comparison, a circuit breaker ofsimilar current rating:v which passes 1.05 In must not trip in lessthan one hour; andv when passing 1.25 In it must trip in onehour, or less(25% overload for up to one hour in the worstcase).

gM fuses require a separate overloadrelay, as described in the note at theend of sub-clause 2.1.

I

t

Inf


fuse-blowncurve

I2

1h.

fig. H2-12: zones of fusing and non-fusingfor gG and gM fuses.

class rated current* conventional non- conventional fusing conventionalIn (A) fusing current current If time h

Inf I2gG In i 4 A 1.5 In 2.1 In 1gM 4 < In < 16 A 1.5 In 1.9 In 1

16 < In i 63 A 1.25 In 1.6 In 163 < In i 160 A 1.25 In 1.6 In 2160 < In i 400 A 1.25 In 1.6 In 3400 < In 1.25 In 1.6 In 4

table H2-13: zones of fusing and non-fusing for LV types gG and gM class fuses (IEC 269-1and 269-2-1).

* Ich for gM fuses

c class aM (motor) fusesThese fuses afford protection against short-circuit currents only and must necessarily beassociated with other switchgear (such asdiscontactors or circuit breakers) in order toensure overload protection < 4 In. They arenot therefore autonomous. Since aM fusesare not intended to protect against low valuesof overload current, no levels of conventionalnon-fusing and fusing currents are fixed. Thecharacteristic curves for testing these fusesare given for values of fault current exceedingapproximately 4 In (see figure H2-14), andfuses tested to IEC 269 must give operatingcurves which fall within the shaded area.Note: the small "arrowheads" in the diagramindicate the current/time "gate" values for thedifferent fuses to be tested (IEC 269).

class aM fuses protect againstshort-circuit currents only, and mustalways be associated with anotherdevice which protects againstoverload.

x In

t

4In

fuse-blowncurve


fig. H2-14: standardized zones of fusingfor type aM fuses (all current ratings).


H2


rated short-circuit breaking

currents

A characteristic of modern cartridge fuses isthat, owing to the rapidity of fusion in the caseof high short-circuit current levels*, a currentcut-off begins before the occurrence of thefirst major peak, so that the fault currentnever reaches its prospective peak value(fig. H2-15).This limitation of current reduces significantlythe thermal and dynamic stresses whichwould otherwise occur, thereby minimizingdanger and damage at the fault position.The rated short-circuit breaking current of thefuse is therefore based on the r.m.s. value ofthe a.c. component of the prospective faultcurrent.No short-circuit current-making rating isassigned to fuses.*for currents exceeding a certain level, depending on thefuse nominal current rating, as shown below in figureH2-15A.

ReminderShort-circuit currents initially contain d.c.components, the magnitude and duration ofwhich depend on the XL/R ratio of the fault-current loop.Close to the source (HV/LV transformer) therelationship I peak / Irms (of a.c. component)immediately following the instant of fault, canbe as high as 2.5 (standardized by IEC, andshown in figure H2-15A).At lower levels of distribution in aninstallation, as previously noted, XL is smallcompared with R and so for final circuits Ipeak / I rms ~ 1.41, a condition whichcorresponds with figure H2-15 above andwith the "n" value corresponding to a powerfactor of 0.95 in table H2-8.

The peak-current-limitation effect occurs onlywhen the prospective r.m.s. a.c. componentof fault current attains a certain level. Forexample, in the above graph the 100 A fusewill begin to cut off the peak at a prospectivefault current (r.m.s.) of 2 kA (a). The samefuse for a condition of 20 kA r.m.s.prospective current will limit the peak currentto 10 kA (b). Without a current-limiting fusethe peak current could attain 50 kA (c) in thisparticular case.As already mentioned, at lower distributionlevels in an installation, R greatlypredominates XL, and fault levels aregenerally low.This means that the level of fault current maynot attain values high enough to cause peak-current limitation. On the other hand, the d.c.transients (in this case) have an insignificanteffect on the magnitude of the current peak,as previously mentioned.

Note on gM fuse ratings.A gM type fuse is essentially a gG fuse, thefusible element of which corresponds to thecurrent value Ich (ch = characteristic) whichmay be, for example, 63 A.This is the IEC testing value, so that its time/current characteristic is identical to that of a63 A gG fuse.This value (63 A) is selected to withstand thehigh starting currents of a motor, the steady-state operating current (In) of which may bein the 10-20 A range.

t

I

0.005s

0.02s

0.01s

prospectivefault-current peakrms value of the a.c.component of theprospective fault currentcurrent peaklimited by the fuse

Tf TaTtc

Tf: fuse pre-arc fusing timeTa: arcing timeTtc: total fault-clearance time

fig. H2-15: current limitation by a fuse.

1 2 5 10 20 50 1001

2

5

10

20

50

100

(a)

(b)

(c)

peak currentcut-offcharacteristiccurves

maximum possible current peak characteristic i.e. 2.5Ir.m.s. (IEC)

160A

100A

50A

nominalfuseratings

prospective fault current (kA) peak

a.c. component of prospectivefault current (kA) r.m.s.

fig. H2-15A: limited peak current versusprospective r.m.s. values of the a.c.component of fault current for LV fuses.

2.1 elementary switching devices (continued)


H2

This means that a physically smaller fusebarrel and metallic parts can be used, sincethe heat dissipation required in normalservice is related to the lower figures(10-20 A).A standard gM fuse, suitable for this situationwould be designated 32M63 (i.e. In M Ich).The first current rating In concerns thesteady-load thermal performance of the fuse-link, while the second current rating (Ich)relates to its (short-time) starting-currentperformance.It is evident that, although suitable for short-circuit protection, overload protection for themotor is not provided by the fuse, and so aseparate thermal-type relay is alwaysnecessary when using gM fuses.The only advantage offered by gM fuses,therefore, when compared with aM fuses, arereduced physical dimensions and slightlylower cost.

2.2 combined switchgear elements

Single units of switchgear do not, in general,fulfil all the requirements of the three basicfunctions, viz: protection, control andisolation.Where the installation of a circuit breaker isnot appropriate (notably where the switchingrate is high, over extended periods)combinations of units specifically designed forsuch a performance are employed.The most commonly-used combinations aredescribed below:

switch and fuse combinations

Two cases are distinguished:c the type in which the operation of one(or more) fuse(s) causes the switch to open.This is achieved by the use of fuses fittedwith striker pins, and a system of switchtripping springs and toggle mechanisms.This type of combination is generally used forcurrent levels exceeding 100 A, and iscommonly associated with a thermal-typeovercurrent relay for overload protection (forwhich the fuses alone may not be suitable).If the switch is classified as AC22 or AC23,and associated with a motor-overload type ofthermal relay, the ensemble, i.e. switch,striker-pin fuses and overload relay, issuitable for the control and protection of amotor circuit,and:c the type in which a non-automatic switch isassociated with a set of fuses in a commonenclosure.In some countries, and in IEC 947-3, theterms "switch-fuse" and "fuse-switch" havespecific meanings, viz:v a switch-fuse comprises a switch (generally2 breaks per pole) on the upstream side ofthree fixed fuse-bases, into which the fusecarriers are inserted (figure H2-17(a)),v a fuse-switch consists of three switchblades each constituting a double-break perphase.These blades are not continuous throughouttheir length, but each has a gap in the centrewhich is bridged by the fuse cartridge. Somedesigns have only a single break per phase,as shown in figures H2-17(a) and (b).

fig. H2-16: symbol for an automatic-tripping switch-fuse, with a thermaloverload relay.

fig. H2-17 (a): symbol for a non-automaticswitch-fuse.

fig. H2-17 (b): symbol for a non-automaticfuse-switch.


H2

The current range for these devices is limitedto 100 A maximum at 400 V 3-phase, whiletheir principal use is in domestic and similarinstallations.To avoid confusion between the first group(i.e. automatic tripping) and the secondgroup, the term "switch-fuse" should bequalified by the adjectives "automatic" or"non-automatic".

fuse - disconnector

+ discontactor

fuse - switch-disconnector

+ discontactor

As previously mentioned, a discontactor doesnot provide protection against short-circuitfaults. It is necessary, therefore, to add fuses(generally of type aM) to perform thisfunction.The combination is used mainly for motor-control circuits, where the disconnector orswitch-disconnector allows safe operationssuch as:c the changing of fuse links (with the circuitisolated);c work on the circuit downstream of thediscontactor (risk of remote closure of thediscontactor).The fuse-disconnector must be interlockedwith the discontactor such that no opening orclosing manœuvre of the fuse-disconnector ispossible unless the discontactor is open(figure H2-18 (a)), since the fuse-disconnector has no load-switching capability.A fuse-switch-disconnector (evidently)requires no interlocking (figure H2-18 (b)).The switch must be of class AC22 or AC23 ifthe circuit supplies a motor.

fig. H2-18 (a): symbol for a fuse-disconnector + discontactor.

fig. H2-18 (b): symbol for a fuse-switch-disconnector + discontactor.

circuit-breaker + contactor

circuit-breaker + discontactor

These combinations are used in remotely-controlled distribution systems in which therate of switching is high, or for control andprotection of a circuit supplying motors.The protection of induction motors isconsidered in chapter J, clause J5.

2.2 combined switchgear elements (continued)



H2

3. choice of switchgear

3.1 tabulated functional capabilities

After having studied the basic functions of LVswitchgear (clause 1, table H2-1) and thedifferent components of switchgear(clause 2), table H2-19 summarizes thecapabilities of the various componentsto perform the basic functions.

isolation control electrical protectionswitchgear functional emergency switching emergency stop switching for overload short-circuit differentialitem (mechanical) mechanical

maintenanceisolator (or cdisconnector)(4)switch (5) c c c (1) c (1) (2) cresidual c c c (1) c (1) (2) c cdevice(RCCB) (5)switch- c c c (1) c (1) (2) cdisconnectorcontactor c c (1) c (1) (2) c c (3)bistable-switch c c (1) c(telerupteur)fuse c c ccircuit c c (1) c (1) (2) c c cbreaker (5)circuit breaker c c c (1) c (1) (2) c c cdisconnector(5)residual c c c (1) c (1) (2) c c c candovercurrentcircuit breaker(RCBO) (5)point of origin of all points where, in general at the at the supply at the supply origin origin origin ofinstallation each for operational incoming circuit point to each point to each of each of each circuits where(general circuit reasons it may to every distribution machine machine circuit circuit the earthingprinciple) be necessary board and/or on the system is

to stop the machine appropriateprocess concerned TN-S, IT, TT

table H2-19: functions fulfilled by different items of switchgear.

(1) Where cut-off of all active conductors is provided(2) It may be necessary to maintain supply to a braking system(3) If it is associated with a thermal relay (the combination is commonly referred to as a "discontactor")(4) In certain countries a disconnector with visible contacts is mandatory at the origin of a LV installation supplied directly from a HV/LV transformer(5) Certain items of switchgear are suitable for isolation duties (e.g. RCCBs according to IEC 1008) without being explicitly marked as such.

3.2 switchgear selection

Software is being used more and more in thefield of optimal selection of switchgear.Each circuit is considered one at a time, anda list is drawn up of the required protectionfunctions and exploitation of the installation,among those mentioned in table H2-19 andsummarized in table H2-1.A number of switchgear combinations arestudied and compared with each otheragainst relevant criteria, with the aim ofachieving:c satisfactory performance;c compatibility among the individual items;from the rated current In to the fault-levelrating Icu;c compatibility with upstream switchgear ortaking into account its contribution;c conformity with all regulations andspecifications concerning safe and reliablecircuit performance.

In order to determine the number of poles foran item of switchgear, reference is made tochapter H1, clause 7, table H1-65.Multifunction switchgear, initially more costly,reduces installation costs and problems ofinstallation or exploitation. It is often foundthat such switchgear provides the bestsolution.


H2

4. circuit breakers

the circuit breaker/disconnector fulfillsall of the basic switchgear functions,while, by means of accessories,numerous other possibilities exist.

As shown in table H2-19 the circuit breaker/disconnector is the only item of switchgearcapable of simultaneously satisfying all thebasic functions necessary in an electricalinstallation.Moreover, it can, by means of auxiliary units,provide a wide range of other functions, forexample: indication (on-off - tripped on fault);undervoltage tripping; remote control… etc.These features make a circuit-breaker/disconnector the basic unit of switchgear forany electrical installation.

functions possible conditionsisolation ccontrol functional c

emergency switching c (with the possibility of a trippingcoil for remote control)

switching-off for mechanical cmaintenance

protection overload cshort-circuit cinsulation faulty c (with differential-current relay)undervoltage c (with undervoltage-trip coil)

remote control c added or incorporatedindication and measurement c (generally optional with an electronic

tripping device)

table H2.20: functions performed by a circuit-breaker/disconnector.

4.1 standards and descriptions

industrial circuit breakers mustconform with IEC 947-1 and 947-2 orother equivalent standards.Corresponding European standardsare presently being developed.Domestic-type circuit breakersshould conform to IEC standard 898,or an equivalent national standard.

standards

For industrial LV installations the relevant IECstandards are, or are due to be:c 947-1: general rules;c 947-2: part 2: circuit breakers;c 947-3: part 3: switches, disconnectors,switch-disconnectors and fuse combinationunits;c 947-4: part 4: contactors and motor-starters;c 947-5: part 5: control-circuit devices andswitching elements;c 947-6: part 6: multiple function switchingdevices;c 947-7: part 7: ancillary equipment.Corresponding European and many nationalstandards are presently in the course ofharmonization with the IEC standards, withwhich they will be in very close agreement.For domestic and similar LV installations, theappropriate standard is IEC 898, or anequivalent national standard.


H2

description

Figure H2-21 shows schematically theprincipal parts of a LV circuit breaker and itsfour essential functions:1 - the circuit-breaking components,comprising the fixed and moving contactsand the arc-dividing chamber.2 - the latching mechanism whichbecomes unlatched by the tripping deviceon detection of abnormal currentconditions.This mechanism is also linked to theoperation handle of the breaker.

3 - a trip-mechanism actuating device:c either: a thermal-magnetic device, in whicha thermally-operated bi-metal strip detects anoverload condition, while an electromagneticstriker pin operates at current levels reachedin short-circuit conditions, or:c an electronic relay operated from currenttransformers, one of which is installed oneach phase.4 - a space allocated to the several types ofterminal currently used for the main power-circuit conductors.

power circuit terminals

trip mechanism and protective devices

latching mechanism

contacts and arc-dividing chamber

fool-proof mechanical indicator

fig. H2-21: principal parts of a circuit breaker.

domestic circuit breakers conformingto IEC 898 and similar nationalstandards perform the basicfunctions of:c isolationc protection against overcurrent.

fig. H2-22: domestic-type circuit breakerproviding overcurrent protection andcircuit isolation features.

some models can be adapted toprovide sensitive detection (30 mA)of earth-leakage current with CBtripping, by the addition of a modularblock, as shown in figure H2-23,while other models (complying withIEC 1009) have this residual-currentfeature incorporated, viz. RCBOs.,and, more recently, IEC 947-2(appendix B) CBRs.

fig. H2-23: domestic-type circuit breakeras above (H2-22) plus protection againstelectric shocks by the addition of amodular block.


H2

4. circuit breakers (continued)

4.1 standards and descriptions (continued)

apart from the above-mentionedfunctions further features can beassociated with the basic circuitbreaker by means of additionalmodules, as shown in figure H2-24;notably remote control and indication(on-off-fault).

fig. H2-24: "Multi 9" system* of LV modular switchgear components.

moulded-case type industrial circuitbreakers conforming to IEC 947-2are now available, which, by meansof associated adaptable blocksprovide a similar range of auxiliaryfunctions to those described above(figure H2-25).

fig. H2-25: example of a modular (Compact NS*) industrial type of circuit breaker capableof numerous auxiliary functions.* Merlin Gerin product.

O-OFF

O-OFF

O-OFFO-OFF

O-OFF

54

32

1

SDE

OF1SD

OF2

OF2SDE

SD

OF1


H2

heavy-duty industrial circuit breakersof large current ratings, conforming toIEC 947-2, have numerous built-incommunication and electronicfunctions (figure H2-26).

fig. H2-26: examples of heavy-duty industrial circuit breakers. The "Masterpact"* providesmany automation features in its tripping module.

These circuit breakers are provided withmeans to adjust protective-device settingsover a wide range, and also with:c a 20 mA output loop;c remote indication contacts;c load indication at the CB.

4.2 fundamental characteristics of a circuit breaker

the fundamental characteristics of acircuit breaker are:c its rated voltage Uec its rated current Inc its tripping-current-level adjustmentranges for overload protection (Ir** orIrth**) and for short-circuit protection(Im)**c its short-circuit current breakingrating (Icu for industrial CBs; Icn fordomestic-type CBs).

rated operational voltage (Ue)

This is the voltage at which the circuit breakerhas been designed to operate, in normal(undisturbed) conditions.Other values of voltage are also assigned tothe circuit breaker, corresponding to disturbedconditions, as noted in sub-clause 4.3.

rated current (In)

This is the maximum value of current that acircuit breaker, fitted with a specified over-current tripping relay, can carry indefinitely atan ambient temperature stated by themanufacturer, without exceeding thespecified temperature limits of the current-carrying parts.Example:A circuit breaker rated at In = 125 A for anambient temperature of 40 °C will beequipped with a suitably calibratedovercurrent tripping relay (set at 125 A).The same circuit breaker can be used athigher values of ambient temperaturehowever, if suitably "derated".Thus, the circuit breaker in an ambienttemperature of 50 °C could carry only 117 Aindefinitely, or again, only 109 A at 60 °C,while complying with the specifiedtemperature limit.Derating a circuit breaker is achievedtherefore, by reducing the trip-current settingof its overload relay, and marking the CBaccordingly. The use of an electronic-type oftripping unit, designed to withstand hightemperatures, allows circuit breakers (deratedas described) to operate at 60 °C (or even at70 °C) ambient.

Note: In for circuit breakers (in IEC 947-2) isequal to Iu for switchgear generally, Iu beingrated uninterrupted current.

frame-size rating

A circuit breaker which can be fitted withovercurrent tripping units of different current-level-setting ranges, is assigned a ratingwhich corresponds with that of the highestcurrent-level-setting tripping unit that can befitted.

* Merlin Gerin products.** Current-level setting values which refer to the current-operated thermal and "instantaneous" magnetic tripping devices forover-load and short-circuit protection.


H2


4.2 fundamental characteristics of a circuit breaker (continued)

overload relay trip-current

setting (Irth or Ir)

Apart from small circuit breakers which arevery easily replaced, industrial circuitbreakers are equipped with removable, i.e.exchangeable, overcurrent-trip relays.Moreover, in order to adapt a circuit breakerto the requirements of the circuit it controls,and to avoid the need to install over-sizedcables, the trip relays are generallyadjustable.The trip-current setting Ir or Irth (bothdesignations are in common use) is thecurrent above which the circuit breaker willtrip. It also represents the maximum currentthat the circuit breaker can carry withouttripping.

That value must be greater than themaximum load current IB, but less than themaximum current permitted in the circuit Iz(see chapter H1, sub-clause 1.3).The thermal-trip relays are generallyadjustable from 0.7 to 1.0 times In, but whenelectronic devices are used for this duty, theadjustment range is greater; typically 0.4 to1 times In.Example (figure H2-27): a circuit breakerequipped with a 320 A overcurrent trip relay,set at 0.9, will have a trip-current setting:Ir = 320 x 0.9 = 288 ANote: for circuit breakers equipped with non-adjustableovercurrent-trip relays, Ir = In.

fig. H2-27: example of a 400 A circuit breaker equipped with a 320 A overload trip unitadjusted to 0.9, to give Ir = 288 A.

short-circuit relay trip-current

setting (Im)

Short-circuit tripping relays (instantaneous orslightly time-delayed) are intended to trip thecircuit breaker rapidly on the occurrence ofhigh values of fault current.Their tripping threshold Im is:c either fixed by standards for domestic typeCBs, e.g. IEC 898, or,c indicated by the manufacturer for industrial-type CBs according to related standards,notably IEC 947-2.

For the latter circuit breakers there exists awide variety of tripping devices which allow auser to adapt the protective performance ofthe circuit breaker to the particularrequirements of a load.

type of pro- overload short-circuittective relay protection protection

domestic thermal- Ir = In low setting standard setting high setting circuitbreakers magnetic type B type C type DIEC 898 3 In i Im < 5 In 5 In i Im < 10 In 10 In i Im < 20 In (1)modular thermal- Ir = In low setting standard setting high settingindustrial (2) magnetic fixed type B or Z type C type D or Kcircuit breakers 3.2 In < fixed < 4.8 In 7 In < fixed < 10 In 10 In < fixed < 14 Inindustrial (2) thermal- Ir = In fixed fixed: Im ≈ 7 to 10 Incircuit breakers magnetic adjustable: adjustable:IEC 947-2 0.7 In i Ir < In - low setting : 2 to 5 In

- standard setting: 5 to 10 Inelectronic long delay short-delay, adjustable

0.4 In i Ir < In 1.5 Ir i Im < 10Irinstantaneous (I) fixedI ≈ 12 to 15 In

table H2.28: tripping-current ranges of overload and short-circuit protective devices for LVcircuit breakers.(1) 50 In in IEC 898, which is considered to be unrealistically high by most European manufacturers (M-G = 10 to 14 In).(2) For industrial use, IEC standards do not specify values. The above values are given only as being those in common use.

overload trip current settingto suit the circuit

Ir

adjustment range

rated current of the tripping unitto suit the circumstances

In0.7 In

circuit-breakerframe-size rating

224 A 288 A 320 A 400 A I


H2

fig. H2-29: performance curve of a circuitbreaker thermal-magnetic protectivescheme.

fig. H2-30 : performance curve of a circuitbreaker electronic protective scheme.

Ir: overload (thermal or short-delay) relay trip-current setting.Im: short-circuit (magnetic or long-delay) relay trip-current setting.I: short-circuit instantaneous relay trip-current setting.PdC: breaking capacity.

isolating feature

A circuit breaker is suitable for isolating acircuit if it fulfills all the conditions prescribedfor a disconnector (at its rated voltage) in therelevant standard (see sub-clause 1.2).In such a case it is referred to as a circuitbreaker-disconnector and marked on its frontface with the symbol

All Multi 9, Compact NS and Masterpact LVswitchgear of Merlin Gerin manufacture is inthis category.

rated short-circuit breaking

capacity (Icu or Icn)

The short-circuit current-breaking rating of aCB is the highest (prospective) value ofcurrent that the CB is capable of breakingwithout being damaged.The value of current quoted in the standardsis the r.m.s. value of the a.c. component ofthe fault current, i.e. the d.c. transientcomponent (which is always present in theworst possible case of short-circuit) isassumed to be zero for calculating thestandardized value. This rated value (Icu) forindustrial CBs and (Icn) for domestic-typeCBs is normally given in kA r.m.s.Icu (rated ultimate s.c. breaking capacity) andIcs (rated service s.c. breaking capacity) aredefined in IEC 947-2 together with a tablerelating Ics with Icu for different categories ofutilization A (instantaneous tripping) and B(time-delayed tripping) as discussed in sub-clause 4.3.Tests for proving the rated s.c. breakingcapacities of CBs are governed by standards,and include:c operating sequences, comprising asuccession of manœuvres, i.e. closing andopening on short-circuit;c current and voltage phase displacement.When the current is in phase with the supplyvoltage (cos ϕ for the circuit = 1), interruptionof the current is easier than that at any otherpower factor.Breaking a current at low lagging* values ofcos ϕ is considerably more difficult toachieve; a zero power-factor circuit being(theoretically) the most onerous case.

the short-circuit current-breakingperformance of a LV circuit breaker isrelated (approximately) to the cos ϕof the fault-current loop. Standardvalues for this relationship have beenestablished in some standards.

In practice, all power-system short-circuit faultcurrents are (more-or-less) at lagging powerfactors, and standards are based on valuescommonly considered to be representative ofthe majority of power systems.In general, the greater the level of faultcurrent (at a given voltage), the lower thepower factor of the fault-current loop, forexample, close to generators or largetransformers.

Table H2-31 below extracted from IEC 947-2relates standardized values of cos ϕ toindustrial circuit breakers according to theirrated Icu.

Icu cos ϕ6 kA < Icu i 10 kA 0.510 kA < Icu i 20 kA 0.320 kA < Icu i 50 kA 0.2550 kA i Icu 0.2

table H2-31: Icu related to power factor(cos ϕ) of fault-current circuit. (IEC 947-2).

c following an open - time delay - close/opensequence to test the Icu capacity of a CB,further tests are made to ensure thatv the dielectric withstand capability;v the disconnection (isolation) performanceandv the correct operation of the overloadprotection,have not been impaired by the test.

I(A)

Im

t (s)

Ir PdC

I(A)

Im

t (s)

Ir PdCI


H2


4.3 other characteristics of a circuit breaker

Familiarity with the following less-importantcharacteristics of LV circuit breakers is,however, often necessary when making afinal choice.

rated insulation voltage (Ui)

This is the value of voltage to which thedielectric tests voltage (generally greaterthan 2 Ui) and creepage distances arereferred.The maximum value of rated operationalvoltage must never exceed that of the ratedinsulation voltage, i.e. Ue i Ui.

rated impulse-withstand voltage

(Uimp)

This characteristic expresses, in kV peak (ofa prescribed form and polarity) the value ofvoltage which the equipment is capable ofwithstanding without failure, under testconditions.For further details see chapter F, clause 2.

category (A or B) and rated

short-time withstand current

(Icw)

As already briefly mentioned (sub-clause 4.2)there are two categories of LV industrialswitchgear, A and B, according to IEC 947-2:c those of category A, for which there is nodeliberate delay in the operation of the"instantaneous" short-circuit magnetic-tripping device (figure H2-32), are generallymoulded-case type circuit breakers, and,c those of category B for which, in order todiscriminate with other circuit breakers on atime basis, it is possible to delay the trippingof the CB, where the fault-current level islower than that of the short-time withstandcurrent rating (Icw) of the CB (figure H2-23).This is generally applied to large open-typecircuit breakers and to certain heavy-dutymoulded-case types. Icw is the maximumcurrent that the B category CB can withstand,thermally and electrodynamically, withoutsustaining damage, for a period of time givenby the manufacturer.

fig. H2-32: category A circuit breaker.

fig. H2-33: category B circuit breaker.

I(A)

Im

t (s)

I(A)

Im

t (s)

I PdCIcw


H2

rated making capacity (Icm)

Icm is the highest instantaneous value ofcurrent that the circuit breaker can establishat rated voltage in specified conditions. In a.c.systems this instantaneous peak value isrelated to Icu (i.e. to the rated breakingcurrent) by the factor k, which depends on thepower factor (cos ϕ) of the short-circuitcurrent loop (as shown in table H2-34).Example: a LV circuit breaker has a ratedbreaking capacity Icu of 100 kA r.m.s.Its rated making capacity Icm will be100 x 2.2 = 220 kA peak.

Icu cos ϕ Icm = kIcu6 kA < Icu i 10 kA 0.5 1.7 x Icu10 kA < Icu i 20 kA 0.3 2 x Icu20 kA < Icu i 50 kA 0.25 2.1 x Icu50 kA i Icu 0.2 2.2 x Icu

table H2.34: relation between ratedbreaking capacity Icu and rated makingcapacity Icm at different power-factorvalues of short-circuit current, asstandardized in IEC 947-2.

rated service short-circuit

breaking capacity (Ics)

The rated breaking capacity (Icu) or (Icn) isthe maximum fault-current a circuit breakercan successfully interrupt without beingdamaged. The probability of such a currentoccurring is extremely low, and in normalcircumstances the fault-currents areconsiderably less than the rated breakingcapacity (Icu) of the CB. On the other hand itis important that high currents (of lowprobability) be interrupted under goodconditions, so that the CB is immediatelyavailable for reclosure, after the faulty circuithas been repaired.It is for these reasons that a new

characteristic (Ics) has been created,expressed as a percentage of Icu, viz: 25, 50,75, 100% for industrial circuit breakers.The standard test sequence is as follows:c O - CO - CO* (at Ics);c tests carried out following this sequenceare intended to verify that the CB is in a goodstate and available for normal service.For domestic CBs, Ics = k Icn. The factor kvalues are given in IEC 898 table XIV.In Europe it is the industrial practice to use ak factor of 100% so that Ics = Icu.Note: O represents an opening operation.CO represents a closing operation followed by an openingoperation.

fault-current limitation

The fault-current limitation capacity of a CBconcerns its ability, more or less effective, inpreventing the passage of the maximumprospective fault-current, permitting only alimited amount of current to flow, as shown infigure H2-35.The current-limitation performance is given bythe CB manufacturer in the form of curves(figure H2-36 diagrams (a) and (b)).c diagram (a) shows the limited peak value ofcurrent plotted against the r.m.s. value of thea.c. component of the prospective faultcurrent ("prospective" fault-current refers tothe fault-current which would flow if the CBhad no current-limiting capability);c limitation of the current greatly reduces thethermal stresses (proportional I2t) and this isshown by the curve of diagram (b) of figureH2-36, again, versus the r.m.s. value of thea.c. component of the prospective faultcurrent.LV circuit breakers for domestic and similarinstallations are classified in certainstandards (notably European StandardEN 60 898). CBs belonging to a class (of

current limiters) have standardized limiting I2tlet-through characteristics defined by thatclass.In these cases, manufacturers do notnormally provide characteristic performancecurves.

in a correctly designed installation,a circuit breaker is never required tooperate at its maximum breakingcurrent Icu.For this reason a new characteristicIcs has been introduced.It is expressed in IEC 947-2 as apercentage of Icu (25, 50, 75, 100%).

many designs of LV circuit breakersfeature a short-circuit currentlimitation capability, whereby thecurrent is reduced and preventedfrom reaching its (otherwise)maximum peak value (figure H2-35).The current-limitation performance ofthese CBs is presented in the form ofgraphs, typified by that shown infigure H2-36, diagram (a).

fig. H2-35: prospective and actualcurrents.

limitedpeakcurrent(kA)

non-

limite

d cu

rrent

char

acte

ristic

s

150

22

prospective a.c.component (r.m.s.)

150 kA

limited peak current (A2 x s)

2.105

4,5.105

prospective a.c.component (r.m.s.)

fig. H2-36: performance curves of a typical LV current-limiting circuit breaker.

(a) (b)

Icc

t

limited current peak

limitedcurrent

tc

prospecticefault-current

prospecticefault-current peak


H2


current limitation reduces boththermal and electrodynamic stresseson all circuit elements through whichthe current passes, therebyprolonging the useful life of theseelements. Furthermore, the limitationfeature allows "cascading"techniques to be used (see 4.5)thereby significantly reducing designand installation costs.

the advantages

of current limitation

The use of current-limiting CBs affordsnumerous advantages:c better conservation of installation networks:current-limiting CBs strongly attenuate allharmful effects associated with short-circuitcurrents;c reduction of thermal effects:conductors (and therefore insulation) heatingis significantly reduced, so that the life ofcables is correspondingly increased;c reduction of mechanical effects:forces due to electromagnetic repulsion arelower, with less risk of deformation andpossible rupture, excessive burning ofcontacts, etc.;c reduction of electromagnetic-interferenceeffects:less influence on measuring instruments andassociated circuits, telecommunicationsystems, etc.These circuit breakers therefore contributetowards an improved exploitation of:c cables and wiring;c prefabricated cable-trunking systems;c switchgear, thereby reducing the ageing ofthe installation.

Example:On a system having a prospective short-circuit current of 150 kA r.m.s., a circuitbreaker limits the peak current to less than10% of the calculated prospective peakvalue, and the thermal effects to less than 1%of those calculated.Cascading of the several levels of distributionin an installation, downstream of a limitingCB, will also result in important economies.The technique of cascading, described insub-clause 4.5 allows, in fact, substantialsavings on switchgear (lower performancepermissible downstream of the limiting CB(s))enclosures, and design studies, of up to 20%(overall).Discriminative protection schemes andcascading are compatible, in the rangeCompact NS*, up to the full short-circuitbreaking capacity of the switchgear.* A Merlin Gerin product.

4.4 selection of a circuit breaker

the choice of a range of circuitbreakers is determined by:the electrical characteristics of theinstallation, the environment, theloads and a need for remote control,together with the type oftelecommunications systemenvisaged.

choice of a circuit breaker

The choice of a CB is made in terms of:c electrical characteristics of the installationfor which the CB is destined;c its eventual environment: ambienttemperature, in a kiosk or switchboardenclosure, climatic conditions, etc.;c short-circuit current breaking and makingrequirements;c operational specifications: discriminativetripping, requirements (or not) for remotecontrol and indication and related auxiliarycontacts, auxiliary tripping coils, connectioninto a local network (communication orcontrol and indication) etc.,

c installation regulations; in particular:protection of persons;c load characteristics, such as motors,fluorescent lighting, LV/LV transformers, etc.Problems concerning specific loads areexamined in chapter J.The following notes relate to the choice of aLV circuit breaker for use in distributionsystems.

choice of rated current in terms

of ambient temperature

The rated current of a circuit breaker isdefined for operation at a given ambienttemperature, in general:c 30 °C for domestic-type CBs;c 40 °C for industrial-type CBs.Performance of these CBs in a differentambient temperature depends principally onthe technology of their tripping units.

ambienttemperature

ambienttemperature

temperature of airsurrounding the circuit breakers

single CBin free air

circuit breakers installedin an enclosure

fig. H2-37: ambient temperature.

4.3 other characteristics of a circuit breaker (continued)


H2

C60a. C60H: curve C. C60N: curves B and C (reference temperature: 30 °C)rating (A) 20 °C 25 °C 30 °C 35 °C 40 °C 45 °C 50 °C 55 °C 60 °C1 1.05 1.02 1.00 0.98 0.95 0.93 0.90 0.88 0.852 2.08 2.04 2.00 1.96 1.92 1.88 1.84 1.80 1.743 3.18 3.09 3.00 2.91 2.82 2.70 2.61 2.49 2.374 4.24 4.12 4.00 3.88 3.76 3.64 3.52 3.36 3.246 6.24 6.12 6.00 5.88 5.76 5.64 5.52 5.40 5.3010 10.6 10.3 10.0 9.70 9.30 9.00 8.60 8.20 7.8016 16.8 16.5 16.0 15.5 15.2 14.7 14.2 13.8 13.520 21.0 20.6 20.0 19.4 19.0 18.4 17.8 17.4 16.825 26.2 25.7 25.0 24.2 23.7 23.0 22.2 21.5 20.732 33.5 32.9 32.0 31.4 30.4 29.8 28.4 28.2 27.540 42.0 41.2 40.0 38.8 38.0 36.8 35.6 34.4 33.250 52.5 51.5 50.0 48.5 47.4 45.5 44.0 42.5 40.563 66.2 64.9 63.0 61.1 58.0 56.7 54.2 51.7 49.2

circuit breakers with uncompensatedthermal tripping units have a trip-current level that depends on thesurrounding temperature.

uncompensated thermal-

magnetic tripping units

Circuit breakers with uncompensated thermaltripping elements have a tripping-current levelthat depends on the surroundingtemperature. If the CB is installed in anenclosure, or in a hot location (boiler room,etc.), the current required to trip the CB onoverload will be sensibly reduced. When thetemperature in which the CB is locatedexceeds its reference temperature, it willtherefore be "derated". For this reason, CBmanufacturers provide tables which indicatefactors to apply at temperatures different to

the CB reference temperature. It may benoted from typical examples of such tables(tables H2-38) that a lower temperature thanthe reference value produces an up-rating ofthe CB.Moreover, small modular-type CBs mountedin juxtaposition, as shown typically in figureH2-24, are usually mounted in a small closedmetal case. In this situation, mutual heating,when passing normal load currents, generallyrequires them to be derated by a factor of 0.8.

NS250N/H/L (reference temperature: 40 °C)rating (A) 40 °C 45 °C 50 °C 55 °C 60 °CTM160D 160 156 152 147 144TM200D 200 195 190 185 180TM250D 250 244 238 231 225

tables H2-38: examples of tables for the determination of derating/uprating factors toapply to CBs with uncompensated thermal tripping units, according to temperature.

ExampleWhat rating (In) should be selected for a CBc protecting a circuit, the maximum loadcurrent of which is estimated to be 34 A;c installed side-by-side with other CBs in aclosed distribution box;c in an ambient temperature of 50 °C.A circuit breaker rated at 40 A would bederated to 35.6 A in ambient air at 50 °C (see

table H2-38). To allow for mutual heating inthe enclosed space, however, the 0.8 factornoted above must be employed, so that,35.6 x 0.8 = 28.5 A, which is not suitable forthe 34 A load.A 50 A circuit breaker would therefore beselected, giving a (derated) current rating of44 x 0.8 = 35.2 A.

compensated thermal-magnetic

tripping units

These tripping units include a bi-metalcompensating strip which allows the overloadtrip-current setting (Ir or Irth) to be adjusted,within a specified range, irrespective of theambient temperature.For example:c in certain countries, the TT system isstandard on LV distribution systems, anddomestic (and similar) installations areprotected at the service position by a circuitbreaker provided by the supply authority.

This CB, besides affording protection againstindirect-contact hazard, will trip on overload;in this case, if the consumer exceeds thecurrent level stated in his supply contract withthe power authority.The circuit breaker (i 60 A) is compensatedfor a temperature range of - 5 °C to + 40 °C.c LV circuit breakers at ratings i 630 A arecommonly equipped with compensatedtripping units for this range (- 5 °C to + 40 °C).


H2


4.4 selection of a circuit breaker (continued)

general note concerning derating

of circuit breakers

It is evident that a CB rated to carry a currentIn at its reference ambient temperature(30 °C) would overheat when carrying thesame current at (say) 50 °C.Since LV CBs are provided with overcurrentprotective devices which (if not compensated)will operate for lower levels of current inhigher ambient temperatures, the CB isautomatically derated by the overload trippingdevice, as shown in the tables H2-38.Where the thermal tripping units aretemperature-compensated, the trippingcurrent level may be set at any valuebetween 0.7 to 1 x In in the ambienttemperature range of - 5 °C to + 40 °C.The reference ambient temperature in thiscase is 40 °C (i.e. on which the rating In isbased).For these compensated units, manufacturers'catalogues generally also give derated valuesof In for ambient temperatures above thecompensated range, e.g. at + 50 °C and+ 60 °C; typically, 95 A at + 50 °C and 90 A at+ 60 °C, for a 100 A circuit breaker.


H2

electronic tripping units are highlystable in changing temperaturelevels.

electronic tripping units

An important advantage with electronictripping units is their stable performance inchanging temperature conditions.However, the switchgear itself often imposesoperational limits in elevated temperatures,

as mentioned in the general note above, sothat manufacturers generally provide anoperating chart relating the maximum valuesof permissible trip-current levels to theambient temperature (figure H2-39).

M25N/H/L iiiii 40 °C 45 °C 50 °C 55 °C 60 °Ccircuit breaker A In (A) 2500 2500 2500 2450 2400

maximum adjustment Ir 1 1 1 0.98 0.96circuit breaker B In (A) 2500 2500 2500 2350 2200

maximum adjustement Ir 1 1 1 0.94 0.88

In (A)coeff.

25001

circuit breaker A24000.96

23500.94

22000.88

20 50 55 6035 40 4525 30

θ °C

circuit breaker B

fig. H2-39: derating of two circuit breakers having different characteristics, according tothe temperature.

selection of an instantaneous, or

short-time-delay, tripping

threshold

Principal charasteristics of magnetic or short-time-delay tripping units. Type classificationaccording to IEC 898. See also table H2-28.

type tripping unit applicationslow setting c sources producing low-short-circuit-current levelstype B (standby generators)

c long lengths of line or cable

standard setting c protection of circuits: general casetype C

high setting c protection of circuits having high initial transienttype D or K current levels (e.g. motors, transformers, resistive

loads)

12 In c protection of motors in association withtype MA discontactors (contactors with overload protection)

I

t

I

t

I

t

I

t

table H2-40: different tripping units, instantaneous or short-time-delayed .


H2



the installation of a LV circuit breakerrequires that its short-circuit breakingcapacity (or that of the CB togetherwith an associated device) be equalto or exceeds the calculatedprospective short-circuit current at itspoint of installation.

selection of a circuit breaker

according to the short-circuit

breaking capacity requirements

The installation of a circuit breaker in a LVinstallation must fulfil one of the two followingconditions:c either have a rated short-circuit breakingcapacity Icu (or Icn) which is equal to orexceeds the prospective short-circuit currentcalculated for its point of installation, orc if this is not the case, be associated withanother device which is located upstream,and which has the required short-circuitbreaking capacity.

In the second case, the characteristics of thetwo devices must be co-ordinated such thatthe energy permitted to pass through theupstream device must not exceed that whichthe downstream device and all associatedcables, wires and other components canwithstand, without being damaged in anyway.This technique is profitably employed in:c associations of fuses and circuit breakers;c associations of current-limiting circuitbreakers and standard circuit breakers.The technique is known as "cascading" (seesub-clause 4.5 of this chapter).

The selection of main and principal circuitbreakersc a single transformerTable C-13 (in chapter C) gives the short-circuit current level on the downstream sideof a commonly-used type of HV/LVdistribution transformer. If the transformer islocated in a consumer's substation, certainnational standards require a LV circuitbreaker in which the open contacts areclearly visible*.

Example (figure H2-41):What type of circuit breaker is suitable for themain circuit breaker of an installation suppliedthrough a 250 kVA HV/LV (400 V) 3-phasetransformer in a consumer's substation?In transformer = 360 AIsc (3-phase) = 8.9 kA.A 400 A CB with an adjustable tripping-unitrange of 250 A-400 A and a short-circuitbreaking capacity (Icu) of 35 kA* would be asuitable choice for this duty.* A type Visucompact NS400N of Merlin Gerin manufactureis recommended for the case investigated.c several transformers in parallel(figure H2-42)v the circuit breakers CBP outgoing from theLV distribution board must each be capable ofbreaking the total fault current from alltransformers connected to the busbars, viz:Isc1 + Isc2 + Isc3,v the circuit breakers CBM, each controllingthe output of a transformer, must be capableof dealing with a maximum short-circuitcurrent of (for example) Isc2 + Isc3 only, for ashort-circuit located on the upstream side ofCBM1.From these considerations, it will be seen thatthe circuit breaker of the smallest transformerwill be subjected to the highest level of faultcurrent in these circumstances, while thecircuit breaker of the largest transformer willpass the lowest level of short-circuit current.v the ratings of CBMs must be chosenaccording to the kVA ratings of the associatedtransformers.

Note: the essential conditions for thesuccessful operation of 3-phase transformersin parallel may be summarized as follows:1. the phase shift of the voltages, primary tosecondary, must be the same in all units to beparalleled.2. the open-circuit voltage ratios, primary tosecondary, must be the same in all units.3. the short-circuit impedance voltage (Zsc%)must be the same for all units. For example, a750 kVA transformer with a Zsc = 6% will

share the load correctly with a 1,000 kVAtransformer having a Zsc of 6%, i.e. thetransformers will be loaded automatically inproportion to their kVA ratings.For transformers having a ratio of kVA ratingsexceeding 2, parallel operation is notrecommended, since the resistance/reactance ratios of each transformer willgenerally be different to the extent that theresulting circulating currrent may overload thesmaller transformer.

the circuit breaker at the output of thesmallest transformer must have ashort-circuit capacity adequate for afault current which is higher than thatthrough any of the other transformerLV circuit breakers (fig. H2-42).

250 kVA20 kV/400 V

VisucompactNS400N

fig. H2-41: example of a transformer in aconsumer's substation.

HV

Tr1

LV

CBMA1

B1

CBP

HV

Tr2

LV

CBMA2

B2

CBP

HV

Tr3

LV

CBMA3

B3

E

fig. H2-42: transformers in parallel.


H2

Table H2-43 indicates, for the most usualarrangement (2 or 3 transformers of equalkVA ratings) the maximum short-circuitcurrents to which main and principal CBs(CBM and CBP respectively, in figure H2-42)are subjected. The table is based on thefollowing hypotheses:c the short-circuit 3-phase power on the HVside of the transformer is 500 MVA;c the transformers are standard 20/0.4 kVdistribution-type units rated as listed;c the cables from each transformer to its LVcircuit breaker comprise 5 metres of single-core conductors;c between each incoming-circuit CBM andeach outgoing-circuit CBP there is 1 metre ofbusbar;c the switchgear is installed in a floor-mounted enclosed switchboard, in anambient-air temperature of 30 °C.Moreover, this table shows selected circuitbreakers of M-G manufacture recommendedfor main and principal circuit breakers in eachcase.

number and minimum S.C. main circuit breakers minimum S.C. rated currentkVA ratings of breaking (CBM) total discrimination breaking cap. In of principal20/0.4 kV capacity of main with out going-circuit of principal circuit breakertransformers CBs (Icu)* kA breakers (CBP) CBs (Icu)* kA (CPB) 250 A2 x 400 14 M08 N1/C 801 N ST 27 NS 250 N3 x 400 27 M08 N1/C 801 N ST 40 NS 250 H2 x 630 22 M10N1/CM1250/C 1001 N 42 NS 250 H3 x 630 43 M10H1/CM1250/C 1001 N 64 NS 250 H2 x 800 24 M12N1/CM1250/C 1251 N 48 NS 250 H3 x 800 48 M12H1/CM1250/C 1251 N 71 NS 250 L2 x 1000 27 M16N1/CM1600 54 NS 250 H3 x 1000 54 M16H2/CM1600 80 NS 250 L2 x 1250 31 M20N1/CM2000 60 NS 250 H3 x 1250 62 M20H1/CM2000 91 NS 250 L2 x 1600 36 M25N1/CM2500 70 NS 250 H3 x 1600 72 M20H2/CM2500H 105 NS 250 L2 x 2000 39 M32H1/CM3200 75 NS 250 L3 x 2000 77 M32H2/CM3200H 112 NS 250 L

table H2-43: maximum values of short-circuit current to be interrupted by main andprincipal circuit breakers (CBM and CBP respectively), for several transformers in parallel.* or Ics in countries where this alternative is practised.

Example: (figure H2-44)c circuit breaker selection for CBM duty:In for an 800 kVA transformer = 1.126 A (at410 V, i.e. no-load voltage) Icu (minimum) =48 kA (from table H2-43), the CBM indicatedin the table is a Compact C1251 N(Icu = 50 kA) (by Merlin Gerin) or itsequivalent;c circuit breaker selection for CBP duty:The s.c. breaking capacity (Icu) required forthese circuit breakers is given in the table(H2-43) as 71 kA.A recommended choice for the three outgoingcircuits 1, 2 and 3 would be current-limitingcircuit breakers types NS 400 L, NS 100 Land NS 250 L respectively (by MG) or theirequivalents. The Icu rating in each case =150 kA.These circuit breakers provide theadvantages of:v absolute discrimination with the upstream(CBM) breakers,v exploitation of the "cascading" technique,with its attendant economy for alldownstream components.

fig H2-44: transformers in parallel.

CBP1

3 Tr800 kVA20 kV/400V

CBM

400 A

CBP2

100 A

CBP3

200 A


H2

short-circuit fault-current levels at anypoint in an installation may beobtained from tables.

Choice of outgoing-circuit CBsand final-circuit CBsccccc use of table H1-40From this table, the value of 3-phase short-circuit current can be determined rapidly forany point in the installation, knowing:v the value of short-circuit current at a pointupstream of that intended for the CBconcerned;v the length, c.s.a., and the composition ofthe conductors between the two points.A circuit breaker rated for a short-circuitbreaking capacity exceeding the tabulatedvalue may then be selected.ccccc detailed calculation of the short-circuitcurrent levelIn order to calculate more precisely the short-circuit current, notably, when the short-circuitcurrent-breaking capacity of a CB is slightlyless than that derived from the table, it isnecessary to use the method indicated inchapter H1 clause 4.ccccc two-pole circuit breakers (for phase andneutral) with one protected pole onlyThese CBs are generally provided with anovercurrent protective device on the phasepole only, and may be used in TT, TN-S andIT schemes. In an IT scheme, however, thefollowing conditions must be respected:v condition (c) of table H1-65 for theprotection of the neutral conductor againstovercurrent in the case of a double fault;v short-circuit current-breaking rating:A 2-pole phase-neutral CB must, byconvention, be capable of breaking on onepole (at the phase-to-phase voltage) thecurrent of a double fault equal to 15% of the3-phase short-circuit current at the point of itsinstallation, if that current is i 10 kA; or 25%of the 3-phase short-circuit current if itexceeds 10 kA;v protection against indirect contact: thisprotection is provided according to the rulesfor IT schemes, as described in chapter Gsub-clause 6.2.ccccc insufficient short-circuit current-breaking ratingIn low-voltage distribution systems itsometimes happens, especially in heavy-dutynetworks, that the Isc calculated exceeds theIcu rating of the CBs available for installation,or system changes upstream result in lower-level CB ratings being exceeded.v solution 1: check whether or notappropriate CBs upstream of the CBsaffected are of the current-limiting type,allowing the principle of cascading (describedin sub-clause 4.5) to be applied;

v solution 2: install a range of CBs having ahigher rating. This solution is economicallyinteresting only where one or two CBs areaffected;v solution 3: associate current-limiting fuses(gG or aM) with the CBs concerned, on theupstream side. This arrangement must,however, respect the following rules:- the fuse rating must be appropriate- no fuse in the neutral conductor, except incertain IT installations where a double faultproduces a current in the neutral whichexceeds the short-circuit breaking rating ofthe CB. In this case, the blowing of theneutral fuse must cause the CB to trip on allphases.

cascading: a particular solution toproblems of CBs insufficiently ratedfor S.C. breaking duty.

associating fuses with CBs avoidsthe need for a fuse in the neutral,except in particular circumstances onsome IT systems.




H2

4.5 coordination between circuit breakers

Preliminary note on the essential functionof current limiting circuit breakersLow-voltage current-limiting CBs exploit theresistance of the short-circuit current arc inthe CB to limit the value of current.An improved method of achieving current-level limitation is to associate a separatecurrent-limiting module (in series) with astandard CB.A contact bar (per phase) in the modulebridges two (specially-designed heavy-duty)contacts, the contact pressure of which isaccurately maintained by springs. Otherrigidly-fixed conductors are arranged in serieswith, and close to the contact bar, such thatwhen current is passed through theensemble, the electromagnetic force tends tomove the contact bar to open its contacts.This occurs at relatively low values of short-circuit current, which then passes through thearcs formed at each contact. The resistanceof the arcs is comparable with systemimpedances at low voltage, so that thecurrent is correspondingly restricted.

Furthermore, the higher the current, the morethe repulsive force on the bar and the greaterthe arc resistance as its path lengthens, i.e.the current magnitude is (to some extent)self-regulating.The circuit breaker is easily able to break theresulting low value of current, particularlysince the power factor of the fault-currentloop is increased by the resistive impedanceof the arcs.When used in a cascading scheme asdescribed below, the tripping of the limitingCB main contacts is briefly delayed, to allowdownstream high-speed circuit breakers toclear the (limited) current, i.e. the current-limiter CB remains closed.The contact bar in the limiter module resetsunder the influence of its pressure springswhen the flow of short-circuit current ceases.Failure of downstream CBs to trip will result inthe tripping of the current-limiting CB, after itsbrief time delay.

cascading

Definition of the cascading techniqueBy limiting the peak value of short-circuitcurrent passing through it, a current-limitingCB permits the use, in all circuits downstreamof its location, of switchgear and circuitcomponents having much lower short-circuitbreaking capacities, and thermal andelectromechanical withstand capabilities thanwould otherwise be the case.Reduced physical size and lowerperformance requirements lead to substantialeconomies and to the simplification ofinstallation work.It may be noted that, while a current-limitingcircuit breaker has the effect on downstreamcircuits of (apparently) increasing the sourceimpedance during short-circuit conditions, ithas no such effect at any other time; forexample, during the starting of a large motor(where a low source impedance is highlydesirable).A new range of Compact* current-limitingcircuit breakers with powerful limitingperformances (namely: NS 100, NS 160,NS 250 and NS 400) is particularlyinteresting.

Conditions of exploitationMost national standards permit use of thecascading technique, on condition that theamount of energy "let through" by the limitingCB is less than that which all downstreamCBs and components are able to withstandwithout damage.In practice this can only be verified for CBs bytests performed in a laboratory. Such testsare carried out by manufacturers who providethe information in the form of tables, so thatusers can confidently design a cascadingscheme based on the combination of circuitbreaker types recommended.By way of an example, table H2-45 indicatesthe possibilities of cascading circuit breakertypes* C 60 and NC 100 when installeddownstream of current-limiting CBsNS 250 N, H or L for a 230/400 V or240/415 V 3-phase installation.

* Merlin Gerin products

the technique of "cascading" usesthe properties of current-limitingcircuit breakers to permit theinstallation of all downstreamswitchgear, cables and other circuitcomponents of significantly lowerperformance than would otherwisebe necessary, thereby simplifyingand reducing the cost of aninstallation.

in general, laboratory tests arenecessary to ensure that theconditions of exploitation required bynational standards are met andcompatible switchgear combinationsmust be provided by themanufacturer.


H2


4.5 coordination between circuit breakers (continued)

Advantages of cascadingThe limitation of current benefits alldownstream circuits that are controlled by thecurrent-limiting CB concerned.The principle is not restrictive, i.e. current-limiting CBs can be installed at any point inan installation where the downstream circuitswould otherwise be inadequately rated.The result is:c simplified short-circuit current calculations;c simplification, i.e. a wider choice ofdownstream switchgear and appliances;c the use of lighter-duty switchgear andappliances, with consequently lower cost;c economy of space requirements, sincelight-duty equipment is generally lessvoluminous.

Short-circuit breaking capacity of theupstream (limiter) CBskA r.m.s.150 NS250L10070 NS250H36 NS250N2522

Short-circuit breaking capacity of thedownstream CBs (benefiting from thecascading technique)kA r.m.s.150 NC100LH

NC100LMA100 NC100LS70 NC100LS NC100L

NC100LHNC100LMA

50 NC100L40 C60L i 40 C60L i 4030 C60H C60N C60N

C60L C60H C60HC60L C60L(50 to 63) (50 to 63)NC100H NC100H

25 C60NNC100H

20 C60a C60a15 C60a

tables H2-45: example of cascadingpossibilities on a 230/400 V or 240/415 V3-phase installation.

discrimination may be absolute orpartial, and based on the principles ofcurrent levels, or time-delays, or acombination of both. A more recentdevelopment is based on theprinciples of logic.A (patented) system by Merlin Gerinexploits the advantages of bothcurrent-limitation and discrimination.

discriminative tripping

(selectivity)

Discrimination is achieved by automaticprotective devices if a fault condition,occurring at any point in the installation, iscleared by the protective device locatedimmediately upstream of the fault, while allother protective devices remain unaffected(figure H2-46).Discrimination between circuit breakers A andB is absolute if the maximum value of short-circuit-current on circuit B does not exceedthe short-circuit trip setting of circuit breakerA. For this condition, B only will trip (figureH2-47).Discrimination is partial if the maximumpossible short-circuit current on circuit Bexceeds the short-circuit trip-current setting ofcircuit breaker A. For this maximum condition,both A and B will trip (figure H2-48).

IscA

IscB

A

B

Iccabsolute discrimination

IccBIrB

Iccpartial discrimination

IccBIrB Ic

B only open A and B opens

fig. H2-46: absolute and partialdiscrimination.


H2

I

t

Irm AIr AIr B

B A

Icc B

Isc downstream of B

I

t

Irm AIr AIr B

B A

A and Bopen

IscA

B only opens

Isc B

fig. H2-48: partial discriminationbetween CBs A and B.

fig. H2-47: absolute discriminationbetween CBs A and B.

1. discrimination based on current levels.This method is realized by setting successiverelay tripping thresholds at stepped levels,from downstream relays (lower settings)towards the source (higher settings).

Discrimination is absolute or partial,according to the particular conditions, asnoted in the above examples.

I

t

Irm AIrm B Isc B

B A

I

t

Isc B

∆t

A

B

A

B

2. discrimination based on stepped timedelays.This method is implemented by adjusting thetime-delayed tripping units, such thatdownstream relays have the shortestoperating times, with progressively longerdelays towards the source.

In the two-level arrangement shown,upstream circuit breaker A is delayedsufficiently to ensure absolute discriminationwith B (for example: Masterpact electronic).

I

t

Irm Adelayed

Isc B

Irm Ainstantaneous

AB

3. discrimination based on a combinationof methods 1 and 2.A mechanical time-delay added to a current-level scheme can improve the overalldiscrimination performance.

Discrimination is absolute ifIsc B < Irm A (instantaneous).The upstream CB has two high-speedmagnetic tripping thresholds:- Irm A (delayed) or a SD* electronic timer- Irm A (instantaneous) standard (Compacttype SA)

* short-delay.

t

IscIrm AIrm B

conventional instantaneousmagnetic-trip characteristic

pressure operatedmagnetic-trip characteristic

table H2-49: summary of methods and components used in order to achieve discriminative tripping.

4. discrimination based on arc-energylevels (Merlin Gerin patent)In the range of short-circuit currents, thissystem provides absolute discriminationbetween two circuit breakers passing thesame fault current. This is achieved by usingcurrent-limiting CBs and initiating CB trippingby pressure-sensitive detectors in the arcing

chambers of the CBs. The heated-airpressure level depends on the energy level ofthe arc, as described in the following pages(figures H2-54 and H2-55).


H2


4.5 coordination between circuit breakers (continued)

current-level discrimination isachieved with stepped current-levelsettings of the instantaneousmagnetic-trip elements.

Current-level discriminationCurrent-level discrimination is achieved withcircuits breakers, preferably limiters, andstepped current-level settings of theinstantaneous magnetic-trip elements.c the downstream circuit breaker is not acurrent-limiter.The discrimination may be absolute or partialfor a short-circuit fault downstream of B, aspreviously noted in 1, above.Absolute discrimination in this situation ispractically impossible because Isc A z Isc B,so that both circuit breakers will generally tripin unison.In this case discrimination is partial, andlimited to the Irm of the upstream circuitbreaker.c the downstream circuit breaker is acurrent limiter.Improvement in discriminative tripping can beobtained by using a current limiter in adownstream location, e.g. for circuitbreaker B.For a short-circuit downstream of B, thelimited level of peak current IB would operatethe (suitably adjusted) magnetic trip unit of B,but would be insufficient to cause circuitbreaker A to trip.Note: All LV breakers (considered here) havesome inherent degree of current limitation,even those that are not classified as current-limiters. This accounts for the curvedcharacteristic shown for the standard circuitbreaker A in figure H2-50.Careful calculation and testing is necessary,however, to ensure satisfactory performanceof this arrangement.c the upstream circuit breaker is high-speed with a short-delay (SD) feature.These circuit breakers are fitted with trip unitswhich include a non-adjustable mechanicalshort-time-delay feature. The delay issufficient to ensure absolute discriminationwith any downstream high-speed CB at anyvalue of s.c. current up to Irms (figure H2-51).

I

I peak

Iscprospective (rms)

current limitationcurve forcircuit breaker(see note)

Isc

A

B

faultupstreamof B

faultdownstreamof B

fig. H2-50: downstream limiting circuitbreaker B.

I

t

Irm Adelayed

A (compact S)

Irm Sinstantaneous

B

only B opens A and B open

fig. H2-51: use of a "selective" circuitbreaker upstream.

Example:circuit breaker A: Compact NS250 N fittedwith a trip unit which includes a SD feature.Ir = 250 A, magnetic trip set at 2,000 Acircuit breaker B: Compact NS100NIr = 100 AThe Merlin Gerin distribution catalogueindicates a discrimination limit of 3,000 A(an improvement over the limit of 2,500 Aobtained when using a standard tripping unit).

Time-based discriminationThis technique requires:c the introduction of "timers" into the trippingmechanisms of CBs;c CBs with adequate thermal and mechanicalwithstand capabilities at the elevated currentlevels and time delays envisaged.Two circuit breakers A and B in series (i.e.passing the same current) are discriminativeif the current-breaking period of downstreambreaker B is less than the non-tripping time ofcircuit breaker A.

Discrimination at several levelsAn example of a practical scheme with (MG)circuit breakers Masterpact (electronicprotection devices).These CBs can be equipped with adjustabletimers which allow 4 time-step selections,such as:c the delay corresponding to a given step isgreater than the total current breaking time ofthe next lower step;

c the delay corresponding to the first step isgreater than the total current-breaking time ofa high-speed CB (type Compact for example)or of fuses (figure H2-52).

I

t

Isc B

A

Isc

B

only B open

current-breakingtime for B

non trippingtime of A

Ir B

fig. H2-52: discrimination by time delay.

discrimination based on time-delayedtripping uses CBs referred to as"selective" (in certain countries).Application of these CBs is relativelysimple and consists in delaying theinstant of tripping of the severalseries-connected circuit breakers in astepped time sequence.


H2

discrimination schemes based onlogic techniques are possible, usingCBs equipped with electronic trippingunits designed for the purpose(Compact, Masterpact by MG) andinterconnected with pilot wires.

Discrimination logicThis discrimination system requires CBsequipped with electronic tripping units,designed for this application, together withinterconnecting pilot wires for data exchangebetween the CBs.With 2 levels A and B (figure H2-53), circuitbreaker A is set to trip instantaneously, unlessthe relay of circuit breaker B sends a signal toconfirm that the fault is downstream of B. Thissignal causes the tripping unit of A to bedelayed, thereby ensuring back-up protectionin the event that B fails to clear the fault, andso on…This system (patented by Merlin Gerin) alsoallows rapid localization of the fault.

A pilot wires

B

fig. H2-53: discrimination logic.

Limitation and discrimination byexploitation of arc energyThe technique of "arc-energy discrimination"(Merlin Gerin patent) is applied on circuitshaving a short-circuit current level u 25 In andensures absolute selectivity between twoCBs carrying the same short-circuit current.Discrimination requires that the energyallowed to pass by the downstream CB (B) isless than that which will cause the upstreamCB (A) to trip (fig. H2-54 (a)).

Operation principleBoth CBs are current limiters, so that theelectromagnetic forces due to a short-circuitdownstream of CB (B) will cause the current-limiting arcing contacts of both CBs to opensimultaneously. The fault current will be verystrongly limited by the resistance of the twoseries arcs. The intense heat of the currentarc in each CB causes a rapid expansion ofthe air in the confined space of the arcingchambers, thereby producing acorrespondingly rapid pressure rise.Above a certain level of current, the pressurerise can be reliably detected and used toinitiate instantaneous tripping.

Discrimination principleIf both CBs include a pressure tripping devicesuitably regulated, then absolutediscrimination between two CBs of differentcurrent ratings can be achieved by settingCB (B) to trip at a lower pressure level thanthat of CB (A) (fig. H2-54). If a short-circuitoccurs downstream of CB (A) but upstreamof CB (B), then the arc resistance of CB (A)only will limit the current. The resulting currentwill therefore be significantly greater than thatoccurring for a short-circuit downstream ofCB (B) (where the two arcs in series cause avery strong limitation, as previously mentioned).The larger current through CB (A) willproduce a correspondingly greater pressure,which will be sufficient to operate itspressure-sensitive tripping unit (diagrams (b)and (c) of fig. H2-54).

As can be seen from figure H2-49 (4), the largerthe short-circuit current, the faster the CB willtrip.Discrimination is assured with this particularswitchgear if:c the ratio of rated currents of the twoCBs u 2.5;c the ratio of the two trip-unit current ratingsis > 1.6, as shown (typically) in figure H2-55.

For overcurrent conditions less than those ofshort-circuits i 25 In, the conventionalprotection schemes are employed, aspreviously described in this chapter.

recently-introduced circuit breakerssuch as Merlin Gerin type NS, usethe principle of arc-energy levels toobtain discrimination.

CB (A) Compact NS

CB (B) Compact NS

Isc = 50 kA

t

CB (A) and CB (B) in series

I

CB (A) only

Isc (limited)

Isc (prospective)

t

CB (B) setting

Pressurein arcingchamber

CB (A) setting

fig. H2-54: arc-energy discriminationprinciples.

NS250NTM260D

NS100NTM100D

CB (A)

CB (B)

fig. H2-55: ratio of rated currents of CBsand of tripping units, must comply withlimits stated in the text, to ensurediscrimination.

(a)

(b)

(c)


H2

4.6 discrimination HV/LV in a consumer's substation

In general the transformer in a consumer'ssubstation is protected by HV fuses, suitablyrated to match the transformer, in accordancewith the principles laid down in IEC 787 andIEC 420, by following the advice of the fusemanufacturer.The basic requirement is that a HV fuse willnot operate for LV faults occurringdownstream of the transformer LV circuitbreaker, so that the tripping characteristiccurve of the latter must be to the left of that ofthe HV fuse pre-arcing curve.This requirement generally fixes themaximum settings for the LV circuit breakerprotection:c maximum short-circuit current-level settingof the magnetic tripping element;c maximum time-delay allowable for theshort-circuit current tripping element.See also Chapter C sub-clause 3.2.7, andAppendix C1, for further details.

63 A

1250 kVA20 kV / 400 V

VisucompactCM 2000set at 1800 A

full-load current1760 A3-phaseshort-circuitcurrent level31.4 kA

fig. H2-56: example.

c short-circuit level at HV terminals oftransformer: 250 MVA;c transformer HL/LV: 1,250 kVA 20/0.4 kV;c HV fuses: 63 A (table C 11);c cabling, transformer - LV circuit breaker:10 metres single-core cables;c LV circuit breaker: Visucompact CM 2000set at 1,800 A (Ir).What is the maximum short-circuit trip currentsetting and its maximum time delayallowable?The curves of figure H2-57 show thatdiscrimination is assured if the short-timedelay tripping unit of the CB is set at:c a level i 6 Ir = 10.8 kA;c a time-delay setting of step O or A.A general policy for HV fuse/LV circuitbreaker discrimination, adopted in somecountries, which is based on standardizedmanufacturing tolerance limits, is mentionedin chapter C sub-clause 3.2.7, and illustratedin figure C-21.Where a transformer is controlled andprotected on the high-voltage side by a circuitbreaker, it is usual to install separate CT- and/or VT- operated relays, which energize ashunt-trip coil of the circuit breaker.Discrimination can be achieved, together withhigh-speed tripping for faults on thetransformer, by using the methods describedin chapter C sub-clause 3.2.

I

t(ms)

step Cstep Bstep A

step 050

1220

10

100200

1000

CM 2000set at1800 A

1800 AIr

Isc maxi31,4 kA

10 kA0,01

minimumpre-arcingcurve for 63 A HVfuses (currentreferred to thesecondary sideof the transformer)

1Ir

4Ir

6Ir

8Ir

fig. H2-57: curves of HV fuses and LVcircuit breaker.


particular supply sources and loads

J

Particular supply sources and loads

1. Protection of circuits supplied by an alternator1.1 an alternator on short-circuit J11.2 protection of essential services circuits supplied in emergencies J4from an alternator1.3 choice of tripping units J51.4 methods of approximate calculation J61.5 the protection of standby and mobile a.c. generating sets J9

2. Inverters and UPS (Uninterruptible Power Supplyunits)2.1 what is an inverter? J102.2 types of UPS system J102.3 standards J112.4 choice of a UPS system J122.5 UPS systems and their environment J142.6 putting into service and technology of UPS systems J152.7 earthing schemes J172.8 choice of main-supply and circuit cables, J20and cables for the battery connection2.9 choice of protection schemes J232.10 complementary equipments J24

3. Protection of LV/LV transformers3.1 transformer-energizing in-rush current J253.2 protection for the supply circuit of a LV/LV transformer J253.3 typical electrical characteristics for LV/LV 50 Hz transformers J263.4 protection of transformers with characteristics as tabled in J3-5 above, J26using Merlin Gerin circuit breakers

4. Lighting circuits4.1 service continuity J294.2 lamps and accessories (luminaires) J304.3 the circuit and its protection J314.4 determination of the rated current of the circuit breaker J314.5 choice of control-switching devices J334.6 protection of ELV lighting circuits J344.7 supply sources for emergency lighting J35

5. Asynchronous motors5.1 protective and control functions required J365.2 standards J385.3 basic protection schemes: circuit breaker / contactor / thermal relay J385.4 preventive or limitative protection J415.5 maximum ratings of motors installed for consumers supplied at LV J435.6 reactive-energy compensation (power-factor correction) J43

6. Protection of direct-current installations6.1 short-circuit currents J446.2 characteristics of faults due to insulation failure, and of protective switchgear J456.3 choice of protective device J456.4 examples in determination of the rated current of the circuit breaker J466.5 protection of persons J47

J

particular supply sources and loads - J1

1. protection of circuits supplied by an alternator

a major difficulty encountered whenan installation may be supplied fromalternative sources (e.g. a HV/LVtransformer or a LV generator) is theprovision of electrical protectionwhich operates satisfactorily oneither source. The crux of theproblem is the great difference in thesource impedances; that of thegenerator being much higher thanthat of the transformer, resulting in acorresponding difference in themagnitudes of fault currents.

Most industrial and large commercialelectrical installations include certainimportant loads for which a power supplymust be maintained, in the event that thepublic electricity supply fails:c either, because safety systems are involved(emergency lighting, automatic fire-protectionequipment, smoke dispersal fans, alarms andsignalization, and so on...) or:c because it concerns priority circuits, such

as certain equipment, the stoppage of whichwould entail a loss of production, or thedestruction of a machine tool, etc.One of the current means of maintaining asupply to the so-called “essential” loads, inthe event that other sources fail, is to install adiesel-generator set connected, via achangeover switch, to an emergency-powerstandby switchboard, from which theessential services are fed (figure J1-1).

G

standby supplychange-over switch

essential loadsnon essential loads

HV

LV

fig. J1-1: example of circuits supplied from a transformer or from an alternator.

1.1 an alternator on short-circuit

the establishment of short-circuit

current (fig. J1-2)Apart from the limited magnitude of faultcurrent from a standby alternator, a furtherdifficulty (from the electrical-protection pointof view) is that during the period in whichLV circuit breakers are normally intended tooperate, the value of short-circuit currentchanges drastically.For example, on the occurrence of a short-circuit at the three phase terminals of analternator, the r.m.s. value of current willimmediately rise to a value of 3 In to 5 In*.An interval of 10 ms to 20 ms following theinstant of short-circuit is referred to as the“sub-transient” period, in which the currentdecreases rapidly from its initial value. Thecurrent continues to decrease during theensuing “transient” interval which may last for80 ms to 280 ms depending on the machinetype, size, etc. The overall phenomenon isreferred to as the “a.c. decrement”. Thecurrent will finally stabilize in about

0.5 seconds, or more, at a value whichdepends mainly on the type of excitationsystem, viz:c manual;c automatic(see figure J1-2).Almost all modern generator sets haveautomatic voltage regulators, compounded tomaintain the terminal voltage sensiblyconstant, by overcoming the synchronousimpedance of the machine as reactive currentdemand changes.This results in an increase in the level of faultcurrent during the transient period to give asteady fault current in the order of 2.5 In to4 In* (figure J1-2).In the (rare) case of manual control of theexcitation, the synchronous impedance of themachine will reduce the short-circuit currentto a value which can be as low as 0.3 In, butis often close to In*.

r.m.s.

0.3 In

3 In

In

subtransientperiod

transientperiod

instantof fault

10 to20 ms

0.1 to0.3 s

alternatorwith automaticvoltage regulator

alternatorwith manualexcitation control

t

fig. J1-2: establishment of short-circuit current for a three-phase short circuit at theterminals of an alternator.* depending on the characteristics of the particular machine.

J

J2 - particular supply sources and loads

1. protection of circuits supplied by an alternator (continued)

1.1 an alternator on short-circuit (continued)

Figure J1-2 shows the r.m.s. values ofcurrent, on the assumption that no d.c.transient components exist. In practice, d.c.components of current are always present tosome degree in at least two phases, beingmaximum when the short-circuit occurs at thealternator terminals.This feature would appear to complicate stillfurther the matter of electrical protection, but,in fact, the d.c. component in each phasesimply increases the r.m.s. values alreadymentioned, so that calculations and tripping-current settings for protective devices basedonly on the a.c. components, as indicatedbelow, will be conservative, i.e. the actualcurrents will always be either equal to orhigher than those calculated.The further the point of short-circuit from thegenerator the lower the fault current, and themore rapidly the transient d.c. componentsdisappear. Furthermore, the a.c. decrementalso becomes negligible when the networkimpedance to the fault position attains ohmicvalues which are high compared with thereactance values of the alternator (since theoverall change in impedance is then relativelysmall).

alternator impedance data

Manufacturers furnish values of the severalimpedances mentioned below. Resistancesare negligibly small compared to thereactances.It can be seen from the constantly-changingvalue of r.m.s. current that the effectivereactance* changes constantly from a lowvalue (sub-transient reactance) to a highvalue (synchronous reactance) in a smoothprogression.The values discussed below are derived fromtest curves and correspond with currentvalues measured at the instant of short-circuit.* An explanation of the significance of thefixed reactance values and how they relate toa smooth variation of current is brieflydescribed in Appendix J1.c the sub-transient reactance x”d isexpressed in % by the manufacturer(analogous to the short-circuit impedancevoltage of a transformer). The ohmic valueX”d is therefore calculated as follows:

X”d (ohms) = x”d Un2 10-5

Pnwhere:x”d is in %Un is in volts (phase/phase)Pn is in kVAc the % transient reactance x’d is given inohms by:

X'd (ohms) = x'd Un2 10-5

Pnc the % zero-phase-sequence reactance x’ois given in ohms by:

X'o (ohms) = x'o Un2 10-5

PnIn the absence of more precise information,the following representative values may beused:x”d = 20% ; x’d = 30 % ; x’o = 6%Pn and Un being, respectively, the rated3-phase power (kVA) and the ratedphase/phase voltage of the alternator (volts).

The sub-transient reactance is used whencalculating the short-circuit current-breakingrating for LV circuit breakers which haveopening times of 20 ms or less, and also forthe electrodynamic stresses to be withstoodby CBs and other components (such asbusbars, cleated single-core cables, etc.).The transient reactance is used whenconsidering the breaking capacity of LV circuitbreakers with an opening time that exceeds20 ms, and also for the thermal withstandcapabilities of switchgear and other systemcomponents.

Remark: from the instant at which the short-circuit is established, the alternator reactancewill rapidly increase. This means that thecurrents calculated from the defined fixedvalues x"d and x'd (for breaking capacity) willalways exceed those that will actually occurat the instant of circuit breaker contactseparation, i.e. there is an inherent safetyfactor incorporated in the current-levelcalculation.These calculations for the circuit breakershort-circuit breaking capacity are based onthe symmetrical a.c. components of currentonly, i.e. no account is taken of the d.c.unidirectional components.For the circuit breaker short-circuit makingcapacity, the d.c. components are crucial, asdiscussed in Chapter C, Sub-clause 1.1(figure C-5).

J


short-circuit current magnitude

at the terminals of an alternator

c the transient 3-phase short-circuit currentat the terminals of an alternator is given by:

Isc = Ig 100* where: x’dIg: rated full-load current of the alternatorx’d = transient reactance per phase of thealternator in %;c when these values are compared withthose for a short-circuit at the LV terminals of

a transformer of equal kVA rating, the currentfrom the alternator will be found to be of theorder of 5 or 6 times less than that from thetransformer. The difference will be evengreater where (as is generally the case) thealternator rating is lower than that of thetransformer.* for CBs with opening time exceeding 20 ms.

non essential loads

630 kVA20 kV/400 VUsc = 4%

A

250 kVA400 VX'd = 30%

essential loads

fig. J1-3: example of an essential services switchboard supplied (in an emergency) from astandby alternator.

Example (figure J1 - 3)What is the value of 3-phase short-circuitcurrent at point A according to the origin ofsupply?Circuit impedances are negligible comparedwith those of the sources.c transformer supply3-phase Isc = 21.5 kA(see table C20 in Chapter C)

c alternator supply

3-phase Isc = Ig x 100 = Pn x 100 x'd eUn x'dwhere: Pn is expressed in kVA

Un is expressed in voltsx’d is expressed in %Isc is expressed in kA

3-phase Isc = 250 x 100 = 1.2 kA ex 400 x 30

J



1.2 protection of essential services circuits supplied in emergencies from an alternator

the difficulty is due to the smallmargin between the rated currentand the short-circuit current of thealternator.

The characteristics (s.c. breaking capacityand range of adjustable magnetic trippingunit) of the CBs protecting the circuits ofessential loads must be defined as describedbelow:

Choice of s.c. breaking capacityThis parameter must always be calculated forthe condition of supply from the transformer,or other “normal” source.Adjustment of magnetic tripping unitsIn practice, the only circuit breakersconcerned are those protecting the essentialservices circuits at the main generaldistribution board.The protection of circuits from localdistribution or sub-distribution boards isalways calibrated at a much lower level thanthose at the main general distribution board,so that, except in unusual cases, adequatefault currents are available from an alternatorto ensure satisfactory protective-gearoperation at these lower levels.Two difficulties have to be overcome:c the first is the need for discrimination ofcircuit protection with the protection schemefor the alternator.For the basic protection requirements of analternator, viz: overload protection, the curveshown in figure J1-4 is representative (seeNote 1).c the second concerns protection of personsagainst electric shock from indirect contact,when the protection depends on theoperation of overcurrent relays (for example,in IT* or TN systems). The operation of theserelays must be assured, whether the supplyis from the alternator or from the transformer(see Note 2).Instantaneous or short-time delay magnetic-relay trip settings of the circuit breakersconcerned must therefore be set to operateat minimum fault levels occurring at theextremity of the circuits they protect, whenbeing supplied from the alternator.

Note 1. Sensitive high-speed protection of analternator against internal faults (i.e.upstream of its CB) is always possible byusing a pilot-wire and current-transformersdifferential scheme of protection, with theadvantage that discrimination with circuitprotection schemes is absolute. The problemof discriminative overload protection (asnoted above) remains, however.A widely-used solution to this problem isprovided by a voltage-controlled overcurrentrelay, which depends on the followingprinciple: short-circuit currents cause muchlower system voltages than overloadcurrents. An inverse-time/current overloadrelay is used having two operating curves,one of which corresponds to that of fig. J1-4,and is effective when system voltage levelsare normal.If the system voltage falls below a pre-setvalue, the relay is automatically switched tooperate much faster and at lower currentlevels than those shown in fig. J1-4.Modern low-setting magnetic tripping units,however, often provide a simpler solution asnoted in 1.3 below.Note 2. Where the level of earth-fault currentis not sufficient, in IT* and TN systems, to tripCBs on overcurrent, the protection againstindirect-contact hazards can be provided byan appropriate use of RCDs, as indicated inChapter G Sub-clause 6.5 Suggestion 2 (forIT circuits) and Sub-clause 5.5. Suggestion 2(for TN circuits).

time (s)

1

32

1.1overload

71012

100

1000

1.2 1.5 2 3 4 5 I/IG

fig. J1-4: overload protection of analternator.

* Two concurrent earth faults on different phases or on one phase and on a neutral conductor, are necessary on IT systems, tocreate an indirect-contact hazard.

J


1.3 choice of tripping units

the calculation of the minimum faultcurrent (in IT or TN schemes) iscomplex. Software packages for thispurpose are available.

calculation of the fault-current

loop impedance (Zs) for IT and

TN systems

The determination of the minimum level ofshort-circuit current, from the calculation ofthe fault-loop impedance Zs (by the sum ofimpedances method) is difficult, mainlybecause of the uncertainly, in a practicalinstallation, of the accuracy of the zero-phase-sequence impedances. Whenconductor routes are known in sufficientdetail, impedances can then be determinedby the use of software, currently availablecommercially. Approximate methods for3-phase and 1-phase short circuits arepresented in Sub-clause 1.4.

types of suitable tripping units

The choice of low-setting magnetic trippingunits will generally be necessary, such asCompact NS* with STR (magnetic-trip shorttime delay is adjustable from 1.5 to 10 Ir) orcircuit breakers Multi 9* curve B (trippingbetween 3 and 5 In).In practice, these CBs (or their equivalents)will always be necessary when the currentrating of the CB is greater than one third ofthe alternator current rating and will, in mostcases, obviate the need for voltage-controlledoverload relays.Switchgear manufacturers often furnishtables showing recommended combinationsof circuit breakers for commonly-usedstandby-generator schemes.* Merlin Gerin products.

characteristics of protectionfor essential-services circuitstype of circuit fault-breaking rating tripping unit adjustment

(FBR)

main FBR > Isc Im or short-delay trip settingcircuits with supply level < the minimum

from transformer fault current at the far endof the circuit when suppliedfrom the alternator(see Note 2 in Sub-clause 1.2)

sub- FBR > Isc check the protectionand final with supply of persons againstcircuits from transformer indirect-contact hazards,

particularly on IT and TNsystems (see Note 2in Sub-clause 1.2)

Isc: 3-ph short-circuit currentIm: magnetic-tripping-relay current setting

loads

B

diesel-generatorprotectioncabinet

power-sourcechangeover switch

fig. J1-5: the protection of essential services circuits.

J



1.4 methods of approximate calculation

An installation on (normal) 630 kVAtransformer supply (figure J1-6) includes anessential-services distribution board whichcan also be supplied from a standby 400 kVAdiesel-alternator set.

What circuit breakers should be installed onthe out-going ways from the essential-services board:c if the installation is TN-earthed?c if the installation is IT-earthed?

sub-distribution board

NS160NTM400D

IB = 92 A70 m35 mm2

PE : 35 mm2

alternatorand dieselprotectionequipmentcabinet

transformer630 kVA20 kV/400 V

alternator400 kVA400 V

NS250NSTR22SE250 A

IB = 220 A100 m120 mm2

PE : 70 mm2

PE

non essential circuitsessential circuits maindistribution board

fig. J1-6: example.

calculation of the minimum level

of 3-phase short-circuit current

Table J1-7 shows the procedure for analternator together with one or severalcircuits.

item of plant R X Z IscmΩ mΩ mΩ kA

alternator Ra X'dcircuit 22.5 L 0.08 x L

Stotal R X R X2 2+ 1.05xVn

R X2 2+table J1-7: procedure for the calculation of 3-phase short-circuit current.

S = c.s.a. in mm2

L = length in metresFor the calculation of cable impedance, referto Chapter H1, Sub-clause 4.2.

J


Consider the 220 A circuit in figure J1-6c alternatorRa = 0

X’d = Un2 x 0.30 = 4002 x 0.30 = 120 mΩ Pn 400c circuit

Rc = 22.5 x 100 = 18.75 mΩ 120Xc = 0.08 x 100 = 8 mΩc application of the method of impedances asindicated in table J1-7;R = Ra + Rc = 0 + 18.75 = 18.75 mΩX = X’d + Xc = 120 + 8 = 128 mΩtotal impedance per phase:

Isc = 1.05 Vn = 1.05 x 230 = 1.87 kA (r.m.s.) Z 0.129

Note: In practice there will always be somemeasure of d.c. transient current in at leasttwo phases, so that the above value willnormally be exceeded during the periodrequired to trip the CB.

Z R X (18.75) (128) 129.4 m2 2 2 2= + = + = Ω

calculation of the minimum level

of 1-phase to earth short-circuit

fault current

Table J1-8 shows the procedure for analternator together with one or severalcircuits.

item of plant R X Z IscmΩ mΩ mΩ kA

alternator Ra 2 X'd + Xo 3

circuit 22.5 L (1 + m) 0.08 x L x 2 Sph

total R X R X2 2+ 1.05xVn

R X2 2+table J1-8: procedure for the calculation of 1-phase to neutral short-circuit current.

For the calculation of cable impedance, referto Chapter H1, Sub-clause 4.2.

Consider the 220 A circuit in figure J1-6c alternatorRa = 0

Xa = (2 x 120 + 4002 x 0.06) x 1 = 88 mΩ 400 3c circuit

Rc = 22.5 x 100 x (1 + 120 / 70) = 50.89 mΩ 120Xc = 0.08 x 100 x 2 = 16 mΩc application of the method of impedances,as for the previous example:R = Ra + Rc = 0 + 50.89 = 50.89 mΩX = Xa + Xc = 88 + 16 = 104 mΩThe total impedance:

and Isc1 (phase/neutral) = 1.05 x 230 = 2.09 kA. 115.8

Z R X 50.89 104 115.8 m2 2 2 2= + = + = Ω

J



1.4 methods of approximate calculation (continued)

maximum permissible setting

of instantaneous or short-time

delay tripping units

c TN schemeOf the two fault conditions considered(3-phase and 1-phase/neutral) the 3-phasefault was found to give the lower short-circuitcurrent. The setting of the protective relaymust therefore be selected to a current levelbelow that calculated.For the 220 A outgoing circuit the trip unitwould be rated at 250 A and adjusted (inprinciple) to Isc/250, i.e. 1,870/250 = 7.4 In.Owing to a ± 20 % manufacturing tolerancehowever, the maximum permissible setting

would be 7.4 = 6.2 In 1.2A tripping unit type TM250D* set at 6 In on aNS250N* circuit breaker (breaking capacity= 36 kA i.e. > 21.5 kA) would be appropriate;c IT schemeIn this case the protection must operate for asecond earth fault occurring before the firstearth fault is cleared. This condition (only)produces indirect-contact hazards on an ITsystem.If the neutral conductor is not distributed, thenthe minimum short-circuit current for thesystem will be the phase-to-phase value(i.e. concurrent earth faults on two differentphases) which is equal to 0.866 Isc(Isc = the 3-phase s.c. current).If the neutral is distributed, the minimum s.c.current occurs when a phase-to-earth faultand a neutral-to-earth fault occurconcurrently, and a protective relay settingequal to 0.5 Isc (phase to neutral) i.e. half thevalue of a phase-to-neutral short-circuitcurrent, is conventionally used to ensurepositive relay operation,v for the case of a non-distributed neutral,the minimum s.c. current =0.5 x 0.866 x 1.87 = 0.81 kAThe tripping unit rated at 250 A will be set at810 x 1 = 2.7 In250 1.2(the factor 1.2 accounting for the ± 20 %manufacturing tolerance for tripping units).A TM250D or a STR22SE tripping unit set at2.5 In would be appropriate,v when the neutral is distributed, theminimum s.c. current relay setting= 0.5 x 2.08 = 1.04 kAThe 250 A tripping unit will be set at1.040 x 1 250 1.2= 3.5 In (the 1.2 factor coveringmanufacturing tolerance, as before)A STR22SE tripping unit, set at 3.0 In wouldbe satisfactory.Note: The foregoing method is based on asimplified application of the followingformulae:➀

Isc (3-phase) = V ph Z1➁

Isc (phase/phase) = eVph Z1+Z2

➂

Isc (phase/earth) = 3 Vph Z1+Z2+Z0

WhereZ1 = positive phase-sequence impedanceZ2 = negative phase-sequence impedanceZ0 = zero phase-sequence impedance

Simplifications:c Z1 is assumed to be equal to Z2 so thatformula ➁ becomeseVph = 0.866 Vph or 0.866 Isc (3-phase) 2 Z1 Z1

c In table J1-8 the calculated cable reactanceassumes that X1 = X2 = X0 for the cable, sothat in formula ③ the total reactance= (X1 + X2 + X0) 1/3 = (3 X1) 1/3 = X1

* Merlin Gerin product.

J


1.5 the protection of standby and mobile a.c. generating sets

Practical guides in certain national standardsclassify generator sets according to threecategories, viz:c permanent installations (as discussed inSub-clauses 1.1 to 1.4);c mobile sets (figure J1-9);c portable power packs (figure J1-10).

mobile sets

These are used mainly to provide temporarysupplies (on construction sites for example)where protection of persons against electricshock must be ensured by the use of RCDswith an operating threshold not exceeding30 mA.

Vigi-compactNS100TM63G30 mA

C32N30 mA

PE

load circuits

non-metallicconduit prividingsupplementaryinsulation

T

PE

fig. J1-9: mobile generating set.

portable power packs

The use of hand-carried power packs by thegeneral public is becoming more and morepopular. When the pack and associatedappliances are not of Class II (i.e. doubleinsulation), 30 mA RCDs are required bymost national standards.

C60N30 mA

T

fig. J1-10: portable power pack with RCDprotection.

J


2. inverters and UPS (Uninterruptible Power Supply units)

2.1 what is an inverter?

An inverter produces an a.c. supply of highquality (i.e. an undistorted sine-wave, freefrom interference) from a d.c. source; itsfunction is the inverse of that of a rectifier(figure J2-1).Its main purpose (when associated with arectifier which provides its input) is to afford ahigh-quality power supply to equipment forwhich the interference and disturbances of anormal power-supply system cannot betolerated (e.g. to computer systems).Power systems are subjected to many kindsof perturbation which adversely affect thequality of supply: atmospheric phenomena(lightning, freezing), accidental faults (short-circuits), industrial parasites, the switching oflarge electric motors (lifts, fluorescent lighting)are among the many causes of poor qualityof supplies.Apart from occasional loss of supply, thedisturbances take the form of more-or-lesssevere voltage dips, high- and low-frequencyparasites, continuous “noise” from

fluorescent-lamp circuits and (normallyundetectable, but totally unacceptable tosensitive electronic systems) of mini-interruptions of several milli-seconds.By the addition of a storage battery at theinput terminals of the inverter (and thereforeacross the output terminals of the associatedrectifier), an elementary UPS system isformed.In normal circumstances, the rectifier suppliesthe load through the inverter, while, at thesame time, a trickle charge from the rectifiermaintains the battery fully charged.A loss of a.c. power supply from thedistribution network would simply result in thebattery automatically maintaining the outputfrom the inverter with no discernableinterruption.

inverter

d.c. source load

sinusoidala.c. output

fig. J2-1: inverter function.

2.2 types of UPS system

there are two main types of UPSsystem:c off-line,c on-line.

Several types of UPS system exist accordingto the degree of protection against power-network “pollution” required, and whethersupply autonomy (automatic standby-supplyon the loss of normal power supply) isspecified, or not. The two most commonly-used types are described below.

An off-line type of UPS system (figure J2-2)is connected in parallel with a supply directfrom the public distribution network, as shownin figure J2-2, and is autonomous, within thecapacity of its battery, on loss of the a.c.power supply. In normal operation the filterimproves the quality of the current while thevoltage is maintained sensibly constant at itsdeclared value by appropriate and automaticregulation within the filter unit.When the tolerance limits are exceeded,including a total loss of supply, a contactor,which carries the normal load, changes overrapidly to the UPS unit (in less than 10 ms)the power then being supplied from thebattery. On the return of normal power supply,the contactor changes back to its originalcondition; the battery then recharges to its fullcapacity.These units are normally of low rating(i 3 kVA) but are capable of passing large

transient currents such as those for motor-starting and switching on of (cold) resistiveloads. The most common use for such unitsis the supply to multi-workstation ITE(information technology equipment)installations, such as cash registers.

sensitiveload

a. c. power supplynetwork

rectifier/charger

battery

inverter filter

F

fig. J2-2: off-line UPS system.

An on-line type of UPS system (figure J2-3)is connected directly between the public a.c.supply network and the load, and has anautonomous capability, the period of whichdepends on the battery capacity and loadmagnitude.The total load passes through the system,which affords a supply of electrical energywithin strict tolerance limits, regardless of thestate of the a.c. power supply network.On loss of the latter, the battery automatically,and without interruption, maintains thepollution-free a.c. supply to the load.This system is equally suitable for small loads(i 3 kVA) or large loads (up to several MVA).

sensitiveload

a.c. power supplynetwork

rectifiercharger

battery

inverter

fig. J2-3: on-line UPS system.

J


types of UPS units, filter plug mains-supply slim-line off-line on-lineconditioners and filters conditioner UPS UPS UPSdiagrams of principle

disturbances consideredtype of network correctivedisturbance measuresHF parasites c c c c cvariations of voltage regulation c c c cautonomy10 to 30 mn (according to battery capacity) c c crated poweri 250 VA c c c c c300 - 1,000 VA c c c c1,000 - 2,500 VA c c c> 2,500 VA c capplications

minimal all micro- micro-informatic highly disturbed a.c.protection sensitive informatic terminals power systems and/or

loads stand-alone PC heavy loads

Other apparatus, not assuring a no-breakperformance, but which protect sensitiveloads from certain disturbances commonlyoccurring on power distribution network,include the following:c the filter-plug which is simply an a.c. plugfor connecting or interconnecting loads, whichhas built-in HF (high-frequency) filters, inorder to reduce HF parasitic interference toacceptable levels. Its principal use is onmicro-informatic stand-alone PCs rated at250 to 1,000 VA, for general office purposes;c the network (or mains) -supply conditioneris a complete system for providing anuncontaminated a.c. power supply, butwithout autonomy, i.e. no provision againstloss of supply from the a.c. distributionnetwork.Its principal functions are to:v filter out HF parasites,v maintain a sensibly-constant voltage level,v isolate (galvanically) the load from the a.c.power network.It is equally applicable to office or industrialsystems which do not require a no-breakstandby supply, up to ratings of 5,000 VA;c the slim-line UPS has integral protectionwith autonomy for each micro-informaticstand-alone PC and its peripherals, and isinstalled immediately under the micro-processor. Two outputs, each with back-upfrom the UPS unit, supply the centralprocessor and screen. Two further outputs,which are filtered, supply other less-sensitiveunits (e.g. the printer). The slim-line UPSbelongs to the class of off-line UPS schemes.

F

F

table J2-4: examples of different possibilities and applications of inverters, in decontamination of supplies and in UPS scheme s.

2.3 standards

The international standard presently coveringsemi-conductor converters is IEC 146-4.

J


2. inverters and UPS (Uninterruptible Power Supply units) (continued)

2.4 choice of a UPS system

The choice of a UPS system is determinedmainly by the following parameters:c rated power, based on:v maximum value of actual estimated kVAdemand,v transitory current peaks (motor starting,energization of resistive loads,transformers...).Note: in order to obtain satisfactorydiscrimination of protective devices for all

types of load, it may be necessary to adjustthe power rating of the UPS system.c voltage levels upstream (input) anddownstream (output) of the UPS unit;c duration of autonomy required (i.e. supplyfrom the battery);c frequencies upstream (input) anddownstream (output) of the UPS unit;c level of availability required.

mains 1

(5) distribution board

(1)(2)

mains 2C/S

(4)

(3) (7)

(6)

UPS

(9)

(8)

fig. J2-5: classical arrangement of a UPS on-line installation, using an inverter.

UPS 1. inverter2. rectifier/charger3. batteries (usual periods of autonomy 10 - 15 - 30 mn - several hours)4. static contactor (see “availability” below)5. isolating transformer, if galvanic isolation from upstream circuits is necessary.6. outgoing ways7. transformer for specific downstream-circuits voltage8. changeover switch9. transformer to match the upstream voltage to that of the consumer.

Note: At first sight, the circuit arrangement infigure J2-5 closely resembles that of theoff-line UPS system (of figure J2-2). In fact,however, it is an on-line system, in which theload is normally passing through circuit 1.The static contactor is open in this situation,but closes automatically if the UPS systembecomes overloaded, or fails for any reason.In such a case, the load will then be suppliedfrom the (reserve) circuit 2. This action is theconverse of that of the off-line scheme.

Conditions will automatically return to normalif the overload, etc. is corrected.

In this arrangement, the voltage output ofthe inverter is always maintained insynchronism with the voltage of the power-supply network (i.e. within close tolerancelimits of magnitude and phase difference)thereby minimizing the disturbance in theevent of “instantaneous” changeover fromcircuit 1 to circuit 2 operation.

power (VA)

The rated power of the UPS unit must besufficient to satisfy the steady load demandas well as loads of a transitory nature. Thedemand will be the sum of the apparent (VA)loads of individual items, for example, theCPU (central processing unit) and willamount to Pa, generally corrected by afactor (1.2 to 2) to allow for futureextensions.However, in order to avoid oversizing of theinstallation, account should be taken of theoverload capacity of the UPS components.For example, inverters manufactured byMerlin Gerin can safely withstand thefollowing overloaded condition:c 1.5 In for 1 minute;c 1.25 In for 10 minutes.

the power rating of a UPS unit musttake account of the peak motor-starting currents, of the possibility offuture extensions to the installation,and of the overload capability of theinverter and other UPS-unitcomponents.

Instantaneous variations of load:these variations occur at times of energizingand de-energizing of one or more items ofload. For an instantaneous change of load upto 100 % of the nominal rating of the UPSunit, the output voltage generally remainsbetween + 10 % and - 8 % of its rated value.

J


Example of a power calculationChoice of a UPS unit suitable for the loadsshown in figure J2-6.

load circuits no.:

1 : 80 kVA

2 : 10 kVA

3 : 20 kVA

4 : 20 kVA

5 : 30 kVA

fig. J2-6: example.

Assumed operating constraints:circuit no. 4 will take a transitory current equalto 4 In for a period of 200 ms when initiallyenergized. This operation will be carried outat least once a day. The peak kVA demand,therefore, represents a supplement (over thesteady-state 20 kVA demand) of 3 x 20 kVA =60 kVA.The remaining circuits require no suchtransitory peak currents. In all cases the kVAvalues cited have taken the load powerfactors into account. Possible futureextensions to the installation are estimated toamount to 20% of the existing load.The maximum steady-state power demandpresently considered is therefore:P = 80 + 10 + 20 + 20 + 30 = 160 kVA.With allowance for extensions (of 20%)= 160 x 1.2 = 192 kVA.With an additional 200 ms peak of (3 x 20)kVA the total amounts to 192 + 60 = 252 kVA.The total of 252 kVA however, includes the60 kVA peak current which is easily absorbedby the 1.5 In overload capability of (a M.G)UPS system, so that the rating of a suitableUPS unit would be 252 x 1/1.5 = 168 kVAfor the nearest standard rating availableabove the calculated value, e.g. 200 kVA.For the choice of suitable protective devices,see Sub-clause 2.9. fig. J2-7: solution to the example.

200 kVA

C/S

availability

A UPS system is generally provided with analternative (unconditioned) emergencysource, a situation which affords a relativelyhigh level of availability.By way of example, a UPS alone has aMTBF (mean time between failures) of50,000 hours.In the usual case, where the supply isdoubled as noted above (mains 1 andmains 2 in figure J2-5) the MTBF obtained isin the range 70,000 to 200,000 hours,depending on the availability of the secondsource.Switching from one source to the other isachieved automatically by a static (solidstate) contactor.Configurations having a higher redundancy,e.g. three UPS units each rated at P/2 tosupply a load of P (figure J2-8) are alsosometimes installed. The calculation of theirlevel of availability can be carried out byspecialists, and the manufacturers are able toquote availability levels, relative to their ownproducts and recommended layouts.

fig. J2-8: 3 UPS P/2 units providing a highlevel of availability of a power rated P.

C/S

P/2

P

P/2

P/2

J



2.5 UPS systems and their environment

UPS system components include themeans to communicate with otherequipments.

UPS units can communicate with otherequipments, notably with IT (informationtechnology) systems, passing dataconcerning the state of the UPS components(static contactor open or closed, and so on...)and receiving orders controlling its function, inorder to:c optimize the protection scheme:the UPS, for example, transmits data(such as: condition normal, supply beingmaintained by the battery, alarm for periodof autonomy almost reached) to the computerit is supplying. The computer deduces theappropriate corrective action, and indicatesaccordingly;c permit remote control:the UPS transmits data concerning the stateof UPS components, together with measuredquantities, to the console of an operator, whois then able to carry out operationalmanœuvres through remote-controlchannels;c supervise (manage) the installation:the consumer (i.e. the “user”) has acentralized management technique facilitywhich allows him to acquire data from theUPS unit(s) which are then stored andanalysed, with anomalies indicated, and thestate of the UPS is presented on a mimicboard or displayed on a screen, and finally toexercise remote control of UPS functions(figures J2-9 to J2-11).

This evolution towards a general compatibilitybetween diverse systems and relatedhardware requires the incorporation of newfunctions in the UPS systems. Thesefunctions can be designed to ensuremechanical and electrical compatibility withother equipments: standard versions are nowprovided with dry contacts and current loops.Interconnection facilities according to thestandards RS 232, RS 422 or RS 485 can beincorporated on request.In fact, certain advanced modules includemodern cards with integral protocole (JBusfor example).Furthermore, they can make use ofspecialized software for automatic checkingand fault diagnosis (e.g. Soft-Monitor on PC)which may be integrated into other systemsof overall supervision (figure J2-10).

fig. J2-9: UPS units can communicate withcentralized system managementterminals.

fig. J2-10: software (e.g. Soft-Monitor)allows remote checking and automaticfault diagnosis of the UPS system.

fig. J2-11: UPS units are readily integrated into centralized management systems.

J


2.6 putting into service and technology of UPS systems

UPS unit for individual stand-alonePCs.

location of UPS units

fig. J2-12: a UPS (slim-line) is easily accommodated under the computer of a stand-alonePC.

fig. J2-13: for large computer installations, the UPS cabinets are generally located in thecomputer room.

UPS system installed in a computerroom.

fig. J2-14: large UPS systems are frequently located in an electrical services room.

UPS cabinets installed in anelectrical services room.

J



2.6 putting into service and technology of UPS systems (continued)

types of battery

Two types of battery are associated withUPS systems.

Maintenance-free sealed unitsThese batteries are used for systems ratedat 250 kVA or less, and provide an autonomyof up to 30 minutes. For certain installations,the natural ventilation of its location isconsidered to be adequate, provided thatthe particular conditions of charging andregulation, together with the characteristicsof the battery, respect the necessaryconstraints. These constraints are defined inthe national standards of some countries(e.g. NF C 15-100, Sub-clause 554 forFrance). To date, there is no IEC equivalentrecommendation, so that consultation withthe battery manufacturer may be advisable.

Non-sealed batteriesThese batteries are generally lead-acid unitsand are used for all large installations.The batteries must be installed in dedicatedbattery rooms, which usually require forced-air ventilation.For certain applications, open-type (i.e. non-sealed) cadmium-nickel batteries arepreferred.

battery location

For any closed location housing batteries,most national standards impose a system ofventilation, forced or natural, which relatesthe renewal rate of air to the size andcharging rate of the battery (or batteries).A recommended air-change rate in cubicmetres per hour can be calculated from theformula 0.05 NI where:N = number of cells in the batteryI = maximum charging-current capability ofthe battery charger (in amperes).

In the case of forced ventilation, the batterycharger must be automatically switched off ifthe fan(s) of the system fail, or if the air-flowis stopped or reduced, for any other reason.For UPS systems of large rating, the batteriesare generally located in specially designedbattery rooms, complying with the relevantlocal standards and regulations.

fig. J2-15: a typical battery room.

J


2.7 earthing schemes

general

In the general case the UPS system is fedfrom two circuits (as shown typically in figureJ2-16) each of which is protected separately;they are referred to as mains 1 and mains 2.Mains 1 is a 3-phase 3-wire circuit connectedto the UPS rectifier/charger input terminals,while mains 2 is a 3-phase 4-wire circuitconnected to the upstream terminals of thestatic contactor. The downstream distributionboard is supplied at 230/400 V.Where other values of voltage are required,adaptor transformers may be employed.

galvanic separation

of the upstream

and downstream circuits

of the UPS system

The measures taken to provide protectionagainst electric shock depend on the earthingscheme, and therefore on the existence, ornot, of galvanic separation of the downstreamcircuits from the upstream circuits.Manufacturers should be ready to provide allthe necessary information.c if there is no separation, then the earthingscheme is evidently identical on both sides ofthe UPS system;

c if there is complete separation between theupstream and downstream sides of the UPSsystem, the earthing schemes upstream anddownstream may be different (or identical).The Technical Notes CT 129 of Merlin Gerinexplain this subject in more detail.

TT/TT scheme

The neutral of the inverter cannot bepermanently connected to earth, asdescribed above, but only temporarily, i.e.when D2 is open in figure J2-16. D2 is a4-pole circuit breaker which breaks theneutral conductor when it is open. The neutralconductor is earthed at the HV/LVtransformer, and so, when D2 opens,contactor C closes automatically to reconnectthe neutral busbar of the LV distribution boardto earth.

General protectionA RCD is installed at each outgoing way ofthe MGDB feeding the UPS system (D1 andD2 in figure J2-16) and discriminationbetween these RCDs and those on theoutgoing ways of the DB downstream of theUPS system, is arranged to ensure themaximum possible continuity of supply.The sensitivity of the RCDs is selectedaccording to the value of earthing resistance(electrode plus earth-wires).Note: Certain versions of RCD are designedto avoid malfunctioning under abnormalconditions (d.c. components of current...) thatare sometimes generated by UPS systems. Itis recommended that the manufacturers ofthe UPS system be consulted concerning thisaspect of their product.

Protection of the d.c. circuits of the UPSsystemc battery protection:most national standards and codes ofpractice, covering battery installations arebased on stringent regulations, which, ifproperly observed, reduces the probability ofa short-circuit fault or an accidental indirectcontact occurring, sufficiently, to consider thatthe circuit from the battery terminals to thecontrolling circuit breaker adequately assuresthe safety of persons. Such will be the case,if:v the battery and all d.c. circuits are in thesame cabinet as the other components of theUPS system, i.e. an equipotential location iscreated,v in the case of a battery location remotefrom the UPS system, class II insulationstandards are respected;c for the remainder of the installation:in particular, the section from the downstreamside of the battery circuit breaker and thejunction of the rectifier output with the inverterinput, where an insulation fault on the d.c.circuits presents a risk, an insulationmonitoring scheme is strongly recommended.A suitable system of permanent surveillanceinjects a low-frequency test current(a XM 200* monitor, as mentioned inMerlin Gerin Technical Notes CT 129, forexample).

Protection of pollution-free output circuitsCircuits supplying socket-outlets will beprotected by RCDs of 30 mA (or less)sensitivity (for example, differential circuitbreakers Multi 9 curve B 30 mA)*. Otheroutgoing ways should be protected by RCDsof suitable sensitivity (in general 300 mA)which must discriminate with the protectionafforded by D1 and 2 (figure J2-16).* Merlin Gerin product.

J



2.7 earthing schemes (continued)

LV distribution board

C/S

UPS

D1

C

RCD30mA

RCD30mA

RCD

RCD

socketsoutletcircuits

D2

mains 2

mains 1

fig. J2-16: TT/TT scheme.

TN-C / TN-S scheme

c the automatic cut-off of supply by indirect-contact hazard protection is achieved in thisscheme by overcurrent relays. Thecalculation of the impedance loop Zshowever, is not possible in this case. Thebasic rule to be observed, is that the short-circuit current from the inverter (which is themaximum it can pass before its internalprotection operates) exceeds that of thetripping threshold of downstream overcurrentprotection.

Circuit breakers with magnetic trip units oflow-setting ranges are suitable for both TN-Cand TN-S schemes. For TN-S installations(only) RCDs of medium sensitivity may alsobe used;c the d.c. section of the UPS system isprotected as previously described for theTT scheme;c protection for the pollution-free outputcircuits will be by 30 mA RCDs for circuitssupplying socket outlets, and by circuitbreakers of low short-circuit tripping settings,previously mentioned.


C/S

UPS

RCD30mA

RCD30mA

socketoutletcircuits

non essentialcircuits

mains 2

mains 1

C

fig. J2-17: TN-C/TN-S scheme.

J


IT/IT scheme

c insulation monitoringThe CIC1 continuous insulation check relay atthe origin of the installation (between theisolated neutral point of the HV/LVtransformer and earth) is automaticallyreplaced by the CIC2 at the output of theinverter, when mains 2 is out of service;c the choice of CICv on the d.c. section of the system, CIC3 usesa very low frequency a.c. current injectionrelay, type Vigilohm XM 200*,v on the a.c. sections, the CIC1, and CIC2

relays are d.c. current-injection relays of typeTR 22A*. In fact, a fault on the d.c. part of thesystem will be detected by CIC1 and CIC2 butthese relays will not operate because theimpedance measurement made by them isnot correct.

CIC current-injection relays operating at verylow frequency (type XM 200* for example)allow correct measurement of the impedance;c reminder of IT system constraintsThe design and operation of an IT systemrequires careful study and exploitation.The advantages of IT operation can only berealized if an in-depth study is completed byclear and concise operating instructions.In particular, the capacitances present in thenetwork (cables and filters on appliances)must be taken into account, and all items ofload must be insulated to withstand phase-to-phase voltage.* a Merlin Gerin product.


C/S

UPSmains 1

mains 2

CIC2

CIC1

fig. J2-18: IT/IT scheme.

complete galvanic separation of

the circuits upstream of the UPS

system from those downstream

Galvanic separation of the upstream anddownstream circuits of the UPS system issometimes required, and is effected byinstalling a 2-winding transformer upstream ofthe static contactor. In this case, the earthingschemes upstream and downstream of theseparation can be different, so that the typeof earthing required for the downstreamcircuits can be created at the outputtransformer of the inverter.

Protection of the d.c. circuits of the UPSsystemThe d.c. circuits of the UPS system areprotected as already described, and theinsulation monitoring relays, if required, areselected as indicated in Sub-clause 2.7 forthe IT/IT scheme.

J



2.8 choice of main-supply and circuit cables, and cables for the battery connection

self-contained UPS units of smallpower ratings are supplied for directconnection, by plugging into theirinput and output sockets.

ready-to-use UPS units

The UPS units for low-power applicationssuch as individual PCs and micro-informaticinstallations are marketed as complete unitsin a metal enclosure, as shown typically infigure J2-19.All internal wiring is factory-installed andadapted to the characteristics of thecomponents.

fig. J2-19: ready-to-use UPS unit.

UPS systems requiring

interconnection of constituent

elements.

For larger UPS installations the battery isgenerally located at some distance from theinverter, and in the case of an off-linearrangement the static contactor and filters (ifinstalled) require interconnection. The cablesizes selected depend on the current level ateach interconnection, as indicated in figureJ2-20, and described below.

in other cases, wiring and cables, forinterconnection of the severalelements of the UPS system, mustbe installed by the consumer’scontractor.

static contactor

rectifier/charger

batterycapacity C10

Ib

inverterload

mains 2

mains 1

Iu

I1

Iu

CS

fig. J2-20: currents to be considered for cable selection.

Calculation of the currents I 1 and Iu

c current Iu is the maximum estimatedutilization current of the load;c current I1 input to the rectifier/charger of theUPS system depends on:v the capacity of the battery (C10) and itscharging rate,v the characteristics of the charger,v the output from the inverter;c the current Ib is the current in the batterycable. These current magnitudes areobtained from the manufacturers of the UPSequipment.

Choice of cablesIn this application the basis of cable selectionis the maximum voltage drop allowable forsatisfactory performance of the load.Preferable values are for this application:c 3% for a.c. circuits;c 1% for d.c. circuits.

Each of these parameters imposes aminimum c.s.a. of conductor. Calculation ofthe c.s.a. of conductors may be carried out asshown in Chapter H1 Clause 2.Merlin Gerin recommends cable sizes to beused with Maxipac and EPS 2000 systems(tables J2-22 to J2-24) in normal conditions,for cable lengths of less than 100 m (voltagedrop < 3 %).Table J2-21 shows the voltage drop for d.c.circuit lengths of less than 100 m of coppercable. That for a.c. cables can be calculatedas described in Chapter H1 Clause 3.

J


nominal current (A) c.s.a. (mm 2)rated of copper-coredpower cables of length < 100 m

circuit 1 with battery (1) circuit 2 circuit 1 circuit 2I1 or or

load load3-phase 400 V 1-phase 230 V 3-phase 1-phase 1-phaseI1 battery I1 battery I1 battery I1 battery Iu 400 V 230 V 230 Vfloating on charge floating on charge

3.5 kVA 18 20 16 16 105 kVA 8.5 10.5 26 28 23 6 16 107.5 kVA 15 19 34 10 1610 kVA 20 24 45.5 10 1615 kVA 30 38 68 10 1620 kVA 40 48 91 10 16

The voltage-drop values in % given in tableJ2-21 correspond to a nominal d.c. voltageof 324 V. For other voltage levels multiplythe table values by a factor equalto the actual battery voltage divided by 324.

c.s.a. mm 2 25 35 50 70 95 120 150 185 240 300In (A) 100 5.1 3.6 2.6 1.9 1.3 1 0.8 0.7 0.5 0.4

125 4.5 3.2 2.3 1.6 1.3 1 0.8 0.6 0.5160 4.0 2.9 2.2 1.6 1.2 1.1 0.8 0.7200 3.6 2.7 2.2 1.6 1.3 1 0.8250 3.3 2.7 2.2 1.7 1.3 1320 3.4 2.7 2.1 1.6 1.3400 3.4 2.8 2.1 1.6500 3.4 2.6 2.1600 4.3 3.3 2.7800 4.2 3.41000 5.3 4.21250 5.3

table J2-21: voltage drop in % of 324 V d.c. for a copper-cored cable.

table J2-22: currents and c.s.a. of copper-cored cables feeding the rectifier, and supplyingthe load for UPS system Maxipac (cable lengths < 100 m).

nominal current (A) c.s.a. (mm 2)rated of copper-coredpower cables of length < 100 m

circuit 1 with battery circuit 2 battery circuit 1 circuit 2 battery3-phase 400 V or 3-phase orI1 load 400 V loadfloating recharging for 400 V Ib 3-phase

standby period of: Iu 400 V10 mn 15 mn 30 mn

10 kVA 19 23 25 25 15.2 27 10 10 1015 kVA 29 36 37 39 22.8 40.5 10 10 1020 kVA 37 49 50 52 30.4 54 16 10 1630 kVA 58 73 76 78 45 81 25 16 2540 kVA 75 97 100 104 60.8 108 35 25 3560 kVA 116 146 151 157 91.2 162 50 35 7080 kVA 151 194 201 209 121.6 216 70 50 95

table J2-23: currents and c.s.a. of copper-cored cables feeding the rectifier, and supplyingthe load for UPS system EPS 2000 (cable lengths < 100 m).Battery cable data are also included.

J



2.8 choice of main-supply and circuit cables, and cables for the battery connection (continued)

nominal current (A)rated power

circuit 1 with battery circuit 2 battery3-phase 400 V - I1 or load Ibfloating recharging for standby period of: 3-phase 400 V

10 mn 15-30 mn Iu40 kVA 70 86 87.6 60.5 10960 kVA 100 123 127 91 16080 kVA 133 158 164 121 212100 kVA 164 198 200 151 255120 kVA 197 240 244 182 317160 kVA 261 317 322 243 422200 kVA 325 395 402 304 527250 kVA 405 493 500 360 658300 kVA 485 590 599 456 790400 kVA 646 793 806 608 1050500 kVA 814 990 1005 760 1300600 kVA 967 1180 1200 912 1561800 kVA 1290 1648 1548 1215 2082

table J2-24: input, output and battery currents for UPS system EPS 5000 (Merlin Gerin).

For a given power rating of a UPS system,these tables indicate the value of inputcurrent I1 to the rectifier/charger when thebattery is on trickle charge (i.e. “floating”) aswell as the load current Iu, together with thec.s.a. of corresponding input and outputcables.The value of I1 when the battery is recharging(following a period in which the load has beentemporarily supplied entirely from the battery)has no influence on the sizing of the cable,due to the short duration of the rechargingcycle. The recharging current has to be takeninto account however, to correctly determinethe upstream protection requirements ofcircuit 1.Example:For a Maxipac UPS system rated at 7.5 kVA3-phase 400 V, I1 = 15 A with the batteryfloating and Iu = 34 A (see table J2-22).The c.s.a. of the corresponding cables are:10 mm2 for the (3-phase) input cable to therectifier/charger,16 mm2 for the (1-phase) output cable to theload.

fig. J2-25: examples of interconnections.

J


2.9 choice of protection schemes

in the choice of protection schemes,it is necessary to take account of thecharacteristics particular to UPSsystems.

In the choice of protection schemes, it isnecessary to take account of thecharacteristics particular to UPS systems: theshort-circuit current from a UPS system isalways very limited, sometimes less than

twice its rated current. Manufacturers carryout tests to ensure a satisfactory co-ordination between the characteristics of theUPS system and the protection afforded byassociated CBs.

choice of circuit breaker ratings

The current ratings (In) of CBs D1, D2, D3and Ddc (figure J2-26) must be chosen such,that:In u I1 for D1 (I1 including the battery re-charging current)In u Iu for D2In u Ide for Ddc

The current rating (In) for each outgoingCB D3 depends on the current rating of theparticular circuit.The currents I1 and Iu for UPS systems ofMerlin Gerin manufacture, are given in tablesJ2-22 to J2-24. The currents Ib are given inthe Merlin Gerin low-voltage distributioncatalogue.

fault-current breaking capacity of

the circuit breakers

Circuit breakers D1 and D2These CBs must have a fault-currentbreaking rating equal to or exceeding thevalue calculated for its location in thenetwork. The calculation is madeconventionally, as previously indicated inChapter H1, Sub-clauses 4.1 and 4.2, forexample.

Circuit breaker D dc

The short-circuit current breaking level for thisCB is always low. In fact, the maximum short-circuit current from a battery is always lessthan 20 times its ampere-hour capacity(battery capacities are indicated in the MerlinGerin low-voltage distribution catalogue).

Circuit breakers D3The very low level of short-circuit currentavailable from the UPS system, gives rise toparticularities concerning the organization ofdiscriminative tripping on the one hand, andprotection against indirect-contact hazards inTN systems, on the other.

c case 1: circuit configuration in which thestatic contactor is closed, but without anyparticular requirement concerning autonomy:the short-circuit current is supplied from thepower network, so that the choice of CBs toensure correct discrimination is determinedby classical methods, previously covered inChapter H2, Sub-clause 4.5;c case 2: circuit configuration without thestatic contactor or with delayed transfer to it,so that discrimination must be achieved byinstantaneous or short time-delay overcurrentprotection, operated by the limited short-circuit current available from the UPS unit,before its internal overcurrent protectionoperates. For Merlin Gerin UPS unitsEPS 5000 or 2000* and Merlin Gerin circuitbreakers, the following conditions must becomplied with:In of a type B circuit breaker

i In of UPS unit 2* In the case of a Maxipac In of a type B

circuit breaker i In of UPS unit 3

example

D2

CS

D1 DdcIb

D3

20 kV / 400 V

630 kVAIsc 22.1 kA

powersystemnetwork

autonomy 10 mn

EPS 5000 of 200 kVAI1

fig. J2-26: example.

Selection of circuit breakers D1 and D2Table J2-24 shows the values of normal-loadcurrents through D1 and D2 respectively, viz:395 A for I1 and 304 A for Iu.The short-circuit current-breaking rating of D1and D2 at their points of installation must be,for such transformers u 22 kA.

Circuit breakers type NS400N* (400 A at40 °C - 36 kA) would be satisfactory;regulated for overload protection (by thermaltripping device) at Irth u 395A for D1and u 304 A for D2.* Merlin Gerin product.

J



2.10 complementary equipments

transformers

A two-winding transformer included on theupstream side of the static contactorof circuit 2 (see figure J2-5) allows:c a change of voltage level when the powernetwork voltage is different to that of the load;c a different arrangement for the neutral onthe load-side winding, from that of the powernetwork.Moreover, such a transformer:c reduces the short-circuit current level onthe secondary, (i.e. load) side compared withthat on the power network side,

c prevents third harmonic currents (andmultiples of them) which may be present onthe secondary side from passing into thepower-system network, providing that theprimary winding is connected in delta.

anti-harmonic filter

The UPS system includes a battery chargerwhich is controlled by commutated thyristorsor transistors. The resulting regularly-chopped current cycles “generate” harmoniccomponents in the power-supply network.These indesirable components are filtered atthe input of the rectifier and for most casesthis reduces the harmonic current levelsufficiently for all practical purposes. Incertain specific cases however, notably invery large installations, an additional filtercircuit may be necessary.

For example, when:c the power rating of the UPS system is largerelative to the HV/LV transformer supplying it;c the LV busbars supply loads which areparticularly sensitive to harmonics;c a diesel (or gas-turbine, etc.) drivenalternator is provided as a standby powersupply.In such cases, the manufacturers of the UPSsystem should be consulted.

communications equipment

Communication with equipment associatedwith informatic systems (see Sub-clause 2.5)may entail the need for suitable facilitieswithin the UPS systems.Such facilities may be incorporated in anoriginal design, or added to existing systemson request.

fig. J2-27: a UPS installation with incorporated communication systems.

J


3. protection of LV/LV transformers

These transformers are generally in the rangeof several hundreds of VA to some hundredsof kVA and are frequently used for:c changing the (LV) voltage level for:v auxiliary supplies to control and indicationcircuits,v lighting circuits (230 V created when theprimary system is 400 V 3-phase 3-wires),c changing the method of earthing for certainloads having a relatively high capacitivecurrent to earth (informatic equipment) orresistive leakage current (electric ovens,industrial-heating processes, mass-cookinginstallations, etc.).LV/LV transformers are generally supplied

with protective systems incorporated, and themanufacturers must be consulted for details.Overcurrent protection must, in any case, beprovided on the primary side. The exploitationof these transformers requires a knowledgeof their particular function, together with anumber of points described below.Note: In the particular cases of LV/LV safetyisolating transformers at extra-low voltage, anearthed metal screen between the primaryand secondary windings is frequentlyrequired, according to circumstances, asrecommended in European StandardEN 60742, and as discussed in detail inSub-clause 3.5 of Chapter G.

3.1 transformer-energizing in-rush current

At the moment of energizing a transformer,high values of transient current (whichincludes a significant d.c. component) occur,and must be taken into account whenconsidering protection schemes. Themagnitude of the current peak depends on:c the value of voltage at the instant ofenergization,c the magnitude and polarity of magnetic flux(if any) existing in the core of the transformer,c characteristics of the load on thetransformer.In distribution-type transformers, the firstcurrent peak can attain a value equal to 10 to15 times the full-load r.m.s. current, but forsmall transformers (< 50 kVA) may reachvalues of 20 to 25 times the nominal full-loadcurrent. This transient current decreasesrapidly, with a time constant θ (see figureJ3-1) of the order of several milli-seconds toseveral tens of milli-seconds.

Î first10 to 25 In

In

θ

I

t

fig. J3-1: transformer-energizing in-rushcurrent.

3.2 protection for the supply circuit of a LV/LV transformer

The protective device on the supply circuit fora LV/LV transformer must avoid the possibilityof incorrect operation due to the magnetizingin-rush current surge, noted above in 3.1. It isnecessary to use therefore:c selective (i.e. slightly time-delayed) circuitbreakers of the type Compact NS STR*(figure J3-2) orc circuit breakers having a very highmagnetic-trip setting, of the typesCompact NS or Multi 9* curve D (figure J3-3).* Merlin Gerin.

t

50 to70 ms

Iinstantaneoustrip

r.m.s. valueof the firstpeak

fig. J3-2: tripping characteristic of aCompact NS STR circuit breaker.

Inr.m.s. valueof the firstpeak

t

I10In 20In

fig. J3-3: tripping characteristic of a circuitbreaker according to standardized type Dcurve (for Merlin Gerin 10 to 14 In).

J


3. protection of LV/LV transformers (continued)

3.2 protection for the supply circuit of a LV/LV transformer (continued)

Example (figure J3-4)A 400 V 3-phase circuit is supplying a125 kVA 400/230 V transformer (In = 180 A)for which the first in-rush current peak canreach 17 In, i.e. 17 x 180 A = 3,067 A.A Compact NS250 circuit breaker with Irsetting of 200 A would therefore be a suitableprotective device.

A particular case: overload protectioninstalled at the secondary side of thetransformerAn advantage of overload protection locatedon the secondary side, is that the short-circuitprotection on the primary side can be set at ahigh value, or alternatively a circuit breakertype MA* may be used. The primary-sideshort-circuit protection setting must, however,be sufficiently sensitive to ensure itsoperation in the event of a short-circuitoccurring on the secondary side of thetransformer (upstream of secondaryprotective devices).* Motor-control circuit breaker, the short-circuit protectiverelay of which is immune to high transient-current peaks, asshown in figure J5-3.

400/230 V125 kVA

3 x 70 mm2

NS250Ntripping unitSTR22SE (Ir = 200)

fig. J3-4: example.

Note: The primary protection is sometimesprovided by fuses, type a M. This practicehas two disadvantages:c the fuses must be largely oversized (atleast 4 times the nominal full-load ratedcurrent of the transformer);c in order to provide isolating facilities on theprimary side, either a load-break switch or acontactor must be associated with the fuses.

3.3 typical electrical characteristics of LV/LV 50 Hz transformers

3-phasekVA rating 5 6.3 8 10 12.5 16 20 25 31.5 40 50 63 80 100 125 160 200 250 315 400 500 630 800no-load losses (W) 100 110 130 150 160 170 270 310 350 350 410 460 520 570 680 680 790 950 1160 1240 1485 1855 2160full-load losses (W) 250 320 390 500 600 840 800 1180 1240 1530 1650 2150 2540 3700 3700 5900 5900 6500 7400 9300 9400 11400 11400s.c. voltage (%) 4.5 4.5 4.5 5.5 5.5 5.5 5.5 5.5 5 5 4.5 5 5 5.5 4.5 5.5 5 5 4.5 6 6 5.5 5.51-phasekVA rating 8 10 12.5 16 20 25 31.5 40 50 63 80 100 125 160no-load losses (W) 105 115 120 140 150 175 200 215 265 305 450 450 525 635full-load losses (W) 400 530 635 730 865 1065 1200 1400 1900 2000 2450 3950 3950 4335s.c. voltage (%) 5 5 5 4.5 4.5 4 4 5 5 4.5 5.5 5 5

table J3-5: typical electrical characteristics of LV/LV 50 Hz transformers.

3.4 protection of transformers with characteristics as tabled in J3-5 above, usingMerlin Gerin circuit breakers

3-phase transformers (400 V primary) circuit breakersP (kVA) In (A) Usc % type trip-unit current

rating (A)/type no.5 7 4.5 C60 / NC100 D or K 2010 14 5.5 C60 / NC100 D or K 3216 23 5.5 C60 / NC100 D or K 6320 28 5.5 C60 / NC100 D or K 6325 35 5.5 NC100 D 8031.5 44 5 NC100 D 8040 56 5 NC100 D 8050 70 4.5 NC100 D 10063 89 5 NS100H/L MA100

NS160H/L STR22SE80 113 5 NS160H/L STR22SE100 141 5.5 NS250N/H/L STR22SE125 176 4.5 NS250N/H/L STR22SE

NS400N/H/L STR23SE160 225 5.5 NS250N/H/L STR22SE

NS400N/H/L STR23SE250 352 5 C801N/H/L STR35SE315 444 4.5 C801N/H/L STR35SE400 563 6 C801N/H/L STR35SE500 704 6 C801NH/L STR35SE

C1001N/H/L STR35SE630 887 5.5 C1001N/H/L STR35SE

C1251N/H STR35SE

table J3-6: protection of 3-phase LV/LV transformers with 400 V primary windings.

J



rating (A)/type no.5 12 4.5 C60 / NC100 D or K 4010 24 5.5 C60 / NC100 D or K 6316 39 5.5 NC100 D 8020 49 5.5 NC100 D 10025 61 5.5 NS100H/L STR22SE31.5 77 5 NS100H/L STR22SE40 97 5 NS100H/L STR22SE50 122 4.5 NS100H/L STR22SE63 153 5 NS250N/H/L STR22SE

NS400N/H/L STR23SE80 195 5 NS250N/H/L STR22SE

NS400N/H/L STR23SE100 244 5.5 NS630N/H/L STR23SE125 305 4.5 C801N/H/L STR35SE160 390 5.5 C801N/H/L STR35SE250 609 5 C801N/H/L STR35SE

C1001N/H/L STR35SE315 767 4.5 C1001N/H/L STR35SE

C1251N/H STR35SE400 974 6 C1251N/H STR35SE




rating (A)/type no.0.1 0.24 13 C60 D or K 10.16 0.39 10.5 C60 D or K 10.25 0.61 9.5 C60 D or K 10.4 0.98 7.5 C60 D or K 20.63 1.54 7 C60 D or K 31 2.44 5.2 C60 D or K 61.6 3.9 4 C60 / NC100 D or K 102 4.88 2.9 C60 / NC100 D or K 102.5 6.1 3 C60 / NC100 D or K 164 9.8 2.1 C60 / NC100 D or K 205 12.2 1.9 C60 / NC100 D or K 326.3 15.4 1.6 C60 / NC100 D or K 408 19.5 5 C60 / NC100 D or K 5010 24 5 C60 / NC100 D or K 6312.5 30 5 C60 / NC100 D or K 6316 39 4.5 NC100 D 8020 49 4.5 NC100 D 10025 61 4.5 NS160H/L STR22SE31.5 77 4 NS160H/L STR22SE40 98 4 NS160H/L STR22SE50 122 4 NS160H/L STR22SE63 154 5 NS250N/H/L STR22SE

NS400N/H/L STR23SE80 195 4.5 NS250N/H/L STR22SE

NS400 STR23SE100 244 5.5 NS630 STR23SE125 305 5 C801N/H/L STR35SE160 390 5 C801N/H/L STR35SE

J



rating (A)/type no.0.1 0.4 13 C60 D or K 10.16 0.7 10.5 C60 D or K 20.25 1.1 9.5 C60 D or K 30.4 1.7 7.5 C60 D or K 40.63 2.7 7 C60 D or K 61 4.2 5.2 C60 / NC100 D or K 101.6 6.8 4 C60 / NC100 D or K 162 8.4 2.9 C60 / NC100 D or K 162.5 10.5 3 C60 / NC100 D or K 204 16.9 2.1 C60 / NC100 D or K 405 21.1 1.9 C60 / NC100 D or K 506.3 27 1.6 C60 / NC100 D or K 638 34 5 NC100 D 8010 42 5 NC100 D 10012.5 53 5 NC100 D 10016 68 4.5 NS160H/L STR22SE20 84 4.5 NS160H/L STR22SE25 105 4.5 NS250N/H/L STR22SE

NS250N/H/L STR22SE31.5 133 4 NS250N/H/L STR22SE40 169 4 NS250N/H/L STR22SE

NS400N/H/L STR23SE50 211 5 NS250N/H/L STR22SE

NS400N/H/L STR23SE63 266 5 NS630N/H/L STR23SE80 338 4.5 C801N/H/L STR35SE100 422 5.5 C801N/H/L STR35SE125 528 5 C801N/H/L STR35SE160 675 5 C801N/H/L STR35SE

C1001N/H/L STR35SE

3. protection of LV/LV transformers (continued)

3.4 protection of transformers with characteristics as tabled in J3-5 above, usingMerlin Gerin circuit breakers (continued)


J


4. lighting circuits

the presence of adequate lightingcontributes to the satety of persons.

The planning and realization of a lightinginstallation requires a sound understanding ofthe materials installed, together withfamiliarity with the rules for safety against firehazards in establishments receiving thepublic.

In fact, the provision of adequate illuminationin the event of fire or other catastrophiccircumstances is of great importance inreducing the likelihood of panic, and inpermitting the necessary safety manœuvresto be carried out.

definitions

Normal lighting refers to the installationdesigned for everyday use.Emergency lighting must ensure easyevacuation of persons from the premisesconcerned, in the event that the normallighting system fails. Furthermore, emergencylighting must be adequate to allow anyparticular safety manœuvres provided in thepremises to be carried out.

Standby lighting is intended to substitutenormal lighting, where the latter fails. Standbylighting permits everyday activities tocontinue more or less normally, depending onthe original design specification, and on theextent of the normal lighting failure. Failure ofthe standby lighting system mustautomatically switch on the emergencylighting system.

emergency lighting is intendedto facilitate the evacuation of personsin case of fire or other panic-causingsituations, when normal lightingsystems may have failed.

4.1 service continuity

continuity of normal lighting servicemust be sufficient, independent ofother supplementary systems.

normal lighting

Regulations governing the minimumrequirements for ERP (EstablishmentsReceiving the Public) in most Europeancountries, are as follows:c installations which illuminate areasaccessible to the public must be controlledand protected independently frominstallations providing illumination to otherareas;c loss of supply on a final lighting circuit (i.e.fuse blown or CB tripped) must not result intotal loss of illumination in an area which iscapable of accommodating more than50 persons;c protection by RCDs (residual currentdifferential devices) must be divided amongstseveral devices (i.e. more than one devicemust be used).

emergency lighting

These schemes include illuminatedemergency exit signs and directionindications, as well as general lighting.c emergency exit indicationsIn areas accommodating more than50 persons, luminous directional indicationsto the nearest emergency exits must beprovided;c general emergency lightingGeneral lighting is obligatory when an areacan accommodate 100 persons or more(50 persons or more in areas below groundlevel).A fault on a lighting distribution circuit mustnot affect any other circuit:v the discrimination of overcurrent-protectionrelays and of RCDs must be absolute, so thatonly the faulty circuit will be cut off,v the installation must be an IT scheme, ormust be entirely class II, i.e. doubly-insulated.

Sub-clause 4.7 describes different kinds ofsuitable power supplies.

in emergency lighting circuits,absolute discrimination betweenprotective devices on the differentcircuits must be provided.

J


4. lighting circuits (continued)

4.2 lamps and accessories (luminaires)

fluorescent tubes

For normal operation a fluorescent tuberequires a ballast and a starter (device forinitiating the luminous discharge).c the ballast is an iron-cored inductor,permanently connected in series with thetube; its function is threefold, viz:v to limit the preheating current during the(brief) starting period,v to provide a pulse of high voltage at the endof the starting period to strike the initial arc,v to stabilize the current through the luminouscolumn (hence the term “ballast”).

The presence of the ballast means that thepower-factor (cos ø) of the circuit is low (ofthe order 0.6) with the correspondingconsumption of reactive energy, which isgenerally metered. For this reason eachfluorescent lamp is normally provided with itsown power-factor-correction capacitor.

c the starter is a switch, which, by breakingthe (electrode-preheating) current passingthrough the ballast, causes a high-voltagetransient pulse to appear across the tube.This causes an arc (in the form of a gaseousdischarge) to be established through thetube. The discharge is then self-sustaining atnormal voltage.The ballast, capacitor and the tube, engenderdisturbances during the periods of starting,steady operation and extinction. Thesedisturbances are analysed in table J4-1below.

switching-on disturbancesc high current peak to chargecapacitor; order of magnitude 10 Infor 1 sec.A number of lamps on one circuitcan result in peaks of 300-400 A for0.5 ms.This can cause a CB to trip, or thewelding of contacts in a contactor. Inpractice, limit each circuit to 8 tubesper contactor;c moderate overload at thebeginning of steady operatingcondition (1.1-1.5 In for 1 sec)according to type of starter.

c no high current peak as notedabove;c same order of moderate overloadat beginning of steady operatingcondition as for the single tubenoted above.This arrangement is recommendedfor difficult cases.

c can generate a current peak atstart;c can cause leakage to earth of HFcurrent (at 30 kHz) via the phaseconductor capacitances to earth.

single-phase fluorescentlamp with its individualp.f. correction capacitor

single-phase twin-tubefluorescent lamp witheach tube having itsown starter and seriesballast. One of the tubeshas a capacitorconnected in series withits ballast. The two setsof equipment areconnected in parallel.The arrangement isknown internationally asa “duo”-circuit luminaire.The capacitor displacesthe phase of the currentthrough its tube, tonullify the flicker effect,as well as correcting theoverall p.f.

fluorescent lamp withHF ballast

Advantages:Energy savings of theorder of 25%.Rapid one-shot start.No flicker orstroboscopic effects.

123

A

B

ballaststarter

starter

table J4-1: analysis of disturbances in fluorescent-lighting circuits.

switching-off disturbancesno particular problems

no particular problems

no particular problems

steady-operating disturbancescirculation of harmonic currents(sinusoidal currents at frequencies equalto whole-number multiples of 50(or 60) Hz:c delta-connected lamps (seeAppendix J2) (3-ph 3-wire 230 V system)

presence of 5th and 7th harmonics at verylow level

c star-connected lamps (3-ph 4-wire400/230 V system)

presence of 3 rd harmonic currents in theneutral, which can reach 70 to 80% of thenominal phase current.In this case, therefore, the c.s.a. of theneutral conductor must equal that of thephase conductors.

123

1

3N

2

J


4.3 the circuit and its protection

dimensions and protection of the

conductors

The maximum currents in the circuits can beestimated using the methods discussed inChapter B.Accordingly, account must be taken of:c the nominal power rating of the lamp andthe ballast;c the power factor.The temperature within the distribution panelalso influences the choice of the protectivedevice (see Chapter H2 Sub-clause 4.4).In general tables are available frommanufacturers to assist in making a choice.

Note: for circuits in which large peak currentsoccur (at times of switching on) and theirmagnitude is such that CB tripping is apossibility, the cable size is chosen after theprotective CB (with an instantaneous tripsetting sufficient to remain closed during thecurrent peaks) has been selected. See theNote following table J4-2.

factor of simultaneity ks

(diversity)

A particular feature of large (e.g. factory)lighting circuits is that the whole load is “on”or “off”, i.e. there is no diversity. Furthermore,even among a number of lighting circuits froma given distribution panel, the factor ks isgenerally near unity.

Consequently, the interior of distributionpanels supplying lighting schemes arefrequently at an elevated temperature, animportant consideration to be taken intoaccount when selecting protective devices.

4.4 determination of the rated current of the circuit breaker

The rated current of a circuit breaker isgenerally chosen according to the rating ofthe circuit conductors it is protecting (in theparticular circumstances in the Note of 4.3above, however, the reverse procedure wasfound to be necessary). The circuit conductorratings are defined by the maximum steadyload current of the circuit.

The following tables allow direct selection ofcircuit breaker ratings for certain particularcases.

power 230 V 1-phase 230 V 3-phase 400 V 3-phase(kW) current rating In (A) current rating In (A) current rating In (A)1 6 3 21.5 10 4 32 10 6 42.5 16 10 43 16 10 63.5 20 10 104 20 16 104.5 25 16 105 25 16 106 32 20 107 32 20 168 40 25 169 50 25 1610 50 32 20

table J4-2: protective circuit breaker ratings for incandescent lamps and resistive-typeheating circuits (see Note below).

Note: at room temperature the filament resistance of a 100 W 230 V incandescent lamp isapproximately 34 ohms. Some milli-seconds after switching on, the filament resistance rises to2302/100 = 529 ohms.The initial current peak at the instant of switch closure is therefore practically 15 times its normaloperating current.A similar (but generally less severe) transient current peak occurs when energizing any resistive-type heating appliance.

J


single-phase distribution 230 Vthree-phase distribution + N : 400 V phase/phasetypes de tube number of luminaires per phaseluminaires rating

(W)single-phase 18 7 14 21 42 70 112 140 175 225 281 351 443 562 703with capacitor 36 3 7 10 21 35 56 70 87 112 140 175 221 281 351

58 2 4 6 13 21 34 43 54 69 87 109 137 174 218duo circuit 2x18= 36 3 7 10 21 35 56 70 87 112 140 175 221 281 351with 2x36= 72 1 3 5 10 17 28 35 43 56 70 87 110 140 175capacitor 2x58= 116 1 2 3 6 10 17 21 27 34 43 54 68 87 109current rating of1-,2-,3 -or 4- pole CBs 1 2 3 6 10 16 20 25 32 40 50 63 80 100


4.4 determination of the rated current of the circuit breaker (continued)

The following table (J4-3) is valid for 230 V and 400 V installations, with or without individualpower-factor correcting capacitors.mercury vapour fluorescent lampsP i 700 W 6 AP i 1000 W 10 AP i 2000 W 16 Ametal-halogen mercury-vapour lampsP 275 W 6 AP 1000 W 10 AP 2000 W 16 Ahigh-pressure sodium discharge lampsP 400 W 6 AP 1000 W 10 A

table J4-3: maximum limit of rated current per outgoing lighting circuit, for high-pressuredischarge lamps.

Calculation for tubes with p.f. capacitor; connected in star

number of tubes per phase = 0.8 C x 0.86 V Pu x 1.25

where: C = current rating of C B, V = phase/neutral voltage, 0.86 = cos ø of circuit, 0.8 = deratingfactor for high temperature in CB housing, 1.25 = factor for watts consumed by ballast,Pu = nominal power rating of tube (W).

three-phase 3 wire system(230 V) phase/phasetypes de tube number of luminaires per phaseluminaires rating

(W)single-phase 18 4 8 12 24 40 64 81 101 127 162 203 255 324 406with capacitor 36 2 4 6 12 20 32 40 50 64 81 101 127 162 203

58 1 2 3 7 12 20 25 31 40 50 63 79 100 126duo circuit 2x18= 36 2 4 6 12 20 32 40 50 64 81 101 127 162 203with 2x36= 72 1 2 3 6 10 16 20 25 32 40 50 63 81 101capacitor 2x58= 116 0 1 1 3 6 10 12 15 20 25 31 39 50 63current rating of2- or 3- pole CBs 1 2 3 6 10 16 20 25 32 40 50 63 80 100

Calculation for tubes with p.f. capacitor; connected in delta

number of tubes per phase = 0.8 C x 0.86 U Pu x 1.25 x e

where: U = phase/phase voltagetables J4-4: current ratings of circuit breakers related to the number of fluorescentluminaires to be protected.

J


4.5 choice of control-switching devices

The advent of switching devices whichcombine the functions of remote control andprotection, of which the remotely-controllableresidual-current circuit breaker is theprototype, simplifies lighting-control circuitsconsiderably, thereby enlarging the scopeand diversity of control schemes.

Certain switching devices include controlcircuitry for operation at ELV (extra-low-voltage, i.e. < 50 V or < 25 V according torequirements); these control circuits beinginsulated for 4,000 V with respect to thepower circuits.The situation at the time of writing issummarized below in table J4-5.

remote-controlmode

point-to-pointremote control

centralizedremote control

point-to-pointand centralizedremote controlcontrol signalsover commun-ications bus

control signalsover time-multiplexingchannels

remotecontrol

bistable switch

contactor

“pilot” bistable switchremote controlledswitch

remotelycontrolledswitch

remote control+ overcurrentprotection

circuitbreakercontrolledby hard-wire system

remotely controlledcircuit breaker overcommunications bus

remote control+ overcurrent protection+ insulation monitoringand protectionresidual current circuitbreaker controlledby hard-wire system

residual current circuitbreaker controlled overcommunications bus

local controldevices

push-button

switch

pushbutton

accordingto type

centralized controldevices

stairway time-switchwith automatic switch-off

automatic photo-electriclighting-control switchesmovement detectors;central clock relaying

function of corresponding switchgear and controlled equipment

remotely controlled static contactor/circuit breaker combination

table J4-5: types of remote control.

J



4.6 protection of ELV lighting circuits

A LV/ELV transformer is often located in aninaccessible position, so that protectioninstalled on the secondary side would beequally difficult to reach.For this reason the protection is commonlyprovided on the primary circuit.The protective device is therefore chosen:c to provide switching control (Multi 9type C CB, or type aM fuses);c to ensure protection against short-circuits.It must therefore be verified that:v in the case of a CB, the minimum value ofshort-circuit current exceeds by a suitablemargin the short-circuit magnetic relay settingIm of the CB concerned,v in the case of fuses it is also necessary toensure that the I2t energy let-through of thefuse(s) at minimum short-circuit current iswell below the level of the thermal withstandcapacity of the circuit conductors,c if necessary, overload protection must beprovided. If the number of lamps on the circuithas been correctly chosen, however,overload protection is not necessary.

Example:The s.c. current Isc2 at the secondaryterminals of a single-phase LV/ELVtransformer is equal toUs where Zs = Us2 x Usc %Zs Pn 100

so that Isc2 = Pn x 100 = 400 x 100 Us x Usc% 12 x 6= 555 A which gives Isc1 = 29 A in theprimary circuit.

Circuit breaker type C trips if the primarycurrent u Im1 = 10 In = 20 A, whichcorresponds to a secondary current of

20 x 230 = 383 A 12The maximum resistance of the ELV (i.e.secondary) circuit* may be deduced fromthese two secondary s.c. currents, viz: 555 Aand 383 A as follows:

Rc = U2 - V2 = 12 - 12 = 0.0313 - 0.0216

Im2 Isc2 383 555= 9.7 mΩ* from the transformer terminals to the ELV distributionboard.

Note: The true value of Rc permitted is,in principle, greater than 9.7 milli-ohms,because the source impedance (i.e. U2/555,will be mainly reactive, not resistive, as(implied) in the example. However,for simplicity, and to automatically providea safety margin under all circumstances,an arithmetic subtraction, as shown,is recommended.

The maximum length of the 12 V circuitbased on 9.7 mΩ will therefore be:Rc (mΩ) x S (mm2) in metres = 9.7 x 6 2 x 22.5 (µΩ.mm) 2 x 22.5for a 6 mm2 copper cable = 1.3 m

It is then necessary to check that this lengthis sufficient to reach the 12 V distributionboard, where the outgoing ways areprotected with other devices. If the length isinsufficient, then an increase in the c.s.a. ofthe conductors, proportional to the increasedlength required, will satisfy the constraint formaximum Rc; for example, a conductor of10 mm2 would allow 1.3 x 10/6 = 2.2 m ofcircuit length in the above case.

230/12 V400 VAUsc = 6%

LV

2 A

ELV

secondarycircuit

fig. J4-6: example.

J


4.7 supply sources for emergency lighting

Supply sources for emergency-lightingsystems must be capable of maintaining thesupply to all lamps in the most unfavourablecircumstances likely to occur, and for a periodjudged necessary to ensure the totalevacuation of the premises concerned,with (in any case) a minimum of one hour.

compatibility between

emergency lighting sources and

other parts of the installation

Emergency-lighting sources must supplyexclusively the circuits installed only foroperation in emergency situations.Standby lighting systems operate to maintainillumination, on failure of normal lightingcircuits (generally in non-emergencycircumstances). However, failure of standbylighting must automatically bring theemergency lighting system into operation.

Central sources for emergency supplies mayalso be used to provide standby supplies,provided that the following conditions aresimultaneously fulfilled:c where there are several sources, the failureof one source must leave sufficient capacityin service to maintain supply to all safetysystems, with automatic load shedding ofnon-essential loads (if necessary);c the failure of one source, or one equipmentconcerned with safety, must leave all othersources and safety equipments unaffected;c any safety equipment must be arranged toreceive supply from any source.

classification of emergency-

lighting schemes

Many countries have statutory regulationsconcerning safety in buildings and areasintended for public gatherings.Classification of such locations leads to thedetermination of suitable types of solutions,authorized for use in emergency-lightingschemes in the different areas.The following four classifications are typical.

Type AThe lamps are supplied permanently andtotally during the presence of the public by asingle central source (battery of storage cells,or a heat-engine-driven generator). Thesecircuits must be independent of any othercircuits (1).

Type BThe lamps are permanently supplied duringthe presence of the public, either:c by a battery to which the lamps arepermanently connected, and which is onpermanent trickle charge from a normallighting source, or,c by a heat-engine-driven generator, thecharacteristics of which also assure suppliesto essential loads within one second (sincethe set is already running and supplying theemergency lighting) in the event of failure ofthe normal power supply, or,c by autonomous units which are normallysupplied and permanently alight from thenormal lighting supply, and which remainalight (for at least one hour), on the loss ofnormal supply, by virtue of a self-containedbattery. The battery is trickle-charged innormal circumstances.These units have fluorescent lamps forgeneral emergency lighting, and fluorescentor incandescent lamps for exit and direction-indicating signs.The circuits for all emergency lamps must beindependent of any other circuits (1).

Type CThe lamps may, or may not, be supplied innormal conditions and, if supplied, may befed from the normal lighting system, or fromthe emergency-lighting supply.c the emergency-lighting batteries must bemaintained on charge from the normal sourceby automatically regulated systems, thatensure a minimum of capacity equal to thefull emergency-lighting load for one hour;c the heat-engine-driven generator sets mustbe capable of automatically picking-up the fullemergency lighting load from a standby(stationary) condition, in less than15 seconds, following the failure of normalsupply.The engine start-up power is provided by abattery which is capable of six startingattempts, or by a system of compressed air.Minimum reserves of energy in the twosystems of start-up must be maintainedautomatically.c failures in the central emergency supplysource must be detected at a sufficientnumber of points and adequately signalled tosupervisory/maintenance personnel;c autonomous units may be of thepermanently-lit type or non-permanently-littype.The circuits for all emergency lamps must beindependent of any other circuits (2).

Type DThis type of emergency lighting compriseshand-carried battery-powered (primary orsecondary cells) at the disposal of servicepersonnel or the public.(1) Circuits for types A and B, in the case of a centralemergency power source, must also be fire-resistant.Conduit boxes, junction sleeves and so on must satisfynational standard heat tests, or the circuits must be installedin protective cable chases, trunking, etc. capable of assuringsatisfactory performance for at least one hour in the event offire.(2) Cable circuits of type C are not required to comply withthe conditions of (1).

J


5. asynchronous motors

the asynchronous (i.e. induction)motor is robust and reliable,and very widely used. 95%of motors installed around the worldare asynchronous. The protectionof these motors is consequentlya matter of great importancein numerous applications.

The consequences of an incorrectly protectedmotor can include the following:c for persons:v asphyxiation due to the blockage of motorventilation,v electrocution due to insulation failure in themotor,v accident due to sticking (contact welding) ofthe controlling contactor;c for the driven machine and the process:v shaft couplings and axles, etc. damageddue to a stalled rotor,v loss of production,v manufacturing time delayed;c for the motor:v motor windings burnt out due to stalledrotor,v cost of dismantling and reinstating orreplacement of motor,v cost of repairs to the motor.

It is, therefore, the safety of persons andgoods, and reliability and availability levelswhich must influence the choice of protectiveequipment.In economic terms, it is the overall cost offailure which must be considered; a penaltywhich is increasingly severe as the size of themotor, and difficulties of access to it increase.Loss of production is a further, and evidentlyimportant factor.

A motor power-supply circuit presents certainconstraints not normally encountered in other(common) distribution circuits, owing to theparticular characteristics, specific to motors,such as:c heavy start-up current (see figure J5-1)which is highly reactive, and can therefore bethe cause of an important voltage drop;c number and frequency of start-upoperations are generally high;c the heavy start-up current means thatmotor overload protective devices must haveoperating characteristics which avoid trippingduring the starting period.

I"IdIn

= 8 to 12 In= 5 to 8 In= nominal motor current

In Id I" I

20 to30 ms

t d1 to 10s

t

fig. J5-1: direct-on-line starting-currentcharacteristics of an induction motor.

specific features of motorperformance influence the power-supply circuits required forsatisfactory operation.

5.1 protective and control functions required

functions to be provided generallyinclude:c basic protective devices,c electronic control equipment,c preventive or limitative protectionequipment.

Functions generally provided are:c basic protection, including:v isolating facility,v manual local and/or remote control,v protection against short-circuits,v protection against overload;c electronic controls consisting of:v progressive “soft-start” motor starter, or,v speed controller;c preventive or limitative protection by meansof:v temperature sensors,v multi-function relays,v permanent insulation-resistance monitor orRCD (residual-current differential device).

Table J5-2 below, shows diverse motor-circuitconfigurations commonly used in LVdistribution boards.

J


basic protection fuse-disconnector circuit breaker* motor circuit contactor circuit+ discontactor + discontactor breaker* + contactor breaker* ACPA(using thermal relay) (using thermal relay)

standards

disconnection(or isolation)

manual remotecontrol control

short-circuitprotection

* circuit breaker includes disconnector capability

overload c large power range c large power range c method is simple c low installation costsprotection c allow all types c avoids need to stock and compact for c no maintenance

of starting schemes fuse cartridges low-power motors c high degree of safetyc a well-proven method c disconnection is and reliabilityc suitable for systems visible in certain cases c suitable for systemshaving high fault c identification of having high fault levelslevels the reason for tripping c long electrical life

refer also to Chapter H2, Sub-clause 2-2 i.e. short circuitor overload

electronic controls

preventive or limitativeprotection devices

progressive “soft-start”starter devicec limitationv current peaks Iv voltage drops Uv mechanical constraintsduring start-up periodc thermal protection isincorporated

speedcontrollerc from 2 to 130 % ofnominal speedc thermal protection isincorporatedc possibility ofcommunication facilities

thermal sensors

Protection against abnormalheating of the motor bythermistance-type sensors in themotor windings, connected toassociated relays.

multi-function relays

Direct and indirect thermal protection against:c the starting period excessively long, orstalled-rotor conditionc imbalance, absence or inversion of phasevoltagesc earth fault or excessive earth-leakage currentc motor running on no-load; motor blockedduring start-upc pre-alarm overheating indication

permanent insulation-to-earth monitor andRCD (residual-current differential relay)

Protection against earth-leakage currentand short-circuits to earth.Signalled indication of need for motormaintenance or replacement.

table J5-2: commonly-used types of LV motor-supply circuits.

J


5. asynchronous motors (continued)

5.2 standards

The international standards coveringmaterials discussed in this Sub-clause are:IEC 947-2, 947-3, 947-4-1, and 947-6-2.These standards are being adopted (oftenwithout any changes) by a number ofcountries, as national standards.

5.3 basic protection schemes: circuit breaker / contactor / thermal relay

functions to be implemented are:c control (start/stop),c isolation (safety duringmaintenance),c protection against short-circuits,c specific protection as noted inSub-clause 5.1Where several different devices areused to provide protection, co-ordination between them isnecessary.

The control and protection of a motor can beprovided by one, two or three devices, whichshare the required functions of:c control (start/stop);c disconnection (isolation) for safety ofpersonnel during maintenance work;c short-circuit protection;c protection specific to the particular motor(but at least thermal relay overcurrentprotection).When these functions are performed byseveral devices, co-ordination between themis essential. In the case of an electrical faultof any kind, none of the devices involvedmust be damaged, except items for whichminor damage is normal in the particularcircumstances, e.g. replaceable arcingcontacts in certain contactors, after a givennumber of service operations, and so on...The kind of co-ordination required dependson the necessary degree of service continuityand on safety levels, etc.

t

In Is

limit of thermal-relay constraint

short-circuit trippingcharacteristicof the circuit breaker(type MA)

l

circuitbreaker

contactor

thermalrelay

câble

motor(nominal current In)

end ofstart-upperiod

ts1 to10 s

20 to30 ms

I" Imagn.

cable thermal-withstandlimit

characteristicsof thermal relay

range 1.05 - 1.20 In

short-circuit-currentbreaking capacities

magneticrelay

circuit breaker only

CB plus contactor (see Note)

among the many possible methodsof protecting a motor, the associationof a circuit breaker incorporating aninstantaneous magnetic trip for short-circuit protection and a contactor witha thermal overload relay* providesmany advantages.

fig. J5-3: tripping characteristics of a circuit breaker (type MA)** andthermal-relay / contactor (1) combination.

AdvantagesThis combination of devices facilitatesinstallation work, as well as operation andmaintenance, by:c the reduction of the maintenance work load:the CB avoids the need to replace blownfuses and the necessity of maintaining astock (of different sizes);c better continuity performance: a motorcircuit can be re-energized immediatelyfollowing the elimination of a fault;c additional complementary devicessometimes required on a motor circuit areeasily accommodated;c tripping of all three phases is assured(thereby avoiding the possibility of “single-phasing”);c full-load current switching possibility (byCB) in the event of contactor failure, e.g.contact welding;

c interlocking;c diverse remote indications;c better protection for the starter for short-circuit currents up to about 30 In (seefigure J5-3).In the majority of cases short-circuit faultsoccur at the motor, so that the current islimited by the cable and the wiring of thestarter (e.g. the direct-acting trip coil of theCB).c possibility of adding RCD:v an RCD of 500 mA sensitivity practicallyeliminates fire risk due to leakage current,v protection against destruction of the motor(short-circuiting of laminations) by the earlydetection of earth-fault currents (300 mA to30 A);c etc.

* The association of an overload relay and a contactor is referred to as a “discontactor” in some countries.** Merlin Gerin.

J


Note: When short-circuit currents are veryhigh, the contacts of some contactors may bemomentarily forced open by electro-magneticrepulsion, so that two sets of contacts (i.e.those of the CB and those of the contactor)are acting in series. The combinationeffectively increases the s.c. current-breakingcapacity above that of the CB alone.

ConclusionThe association circuit breaker / contactor /thermal relay(1) for the control and protectionof motor circuits is eminently appropriatewhere:c the maintenance service for an installationis reduced, which is generally the case intertiary and small-and medium-sizedindustrial enterprises;c the job specification calls forcomplementary functions;c there is an operational requirement for aload-breaking facility in the event of contactwelding of the contactor.(1) a contactor in association with a thermal relay iscommonly referred to as a discontactor.

standardization of the

association of circuit breakers/

discontactors

Categories of contactorThe standard IEC 947-4 gives utilizationcategories which considerably facilitate thechoice of a suitable contactor for a givenservice duty.The utilization categories advise on:c a range of functions for which the contactormay be adapted;c its current breaking and makingcapabilities;c standard test values for expected lifeduration on load, according to its utilization.The following table gives some typicalexamples of the utilization categoriescovered.

utilization application characteristicscategoryAC-1 Non-inductive (or slightly inductive) loads: cos ø u 0.95 (heating, distribution)AC-2 Starting and switching off of slip-ring motorsAC-3 Cage motors: starting, and switching off motors during runningAC-4 Cage motors: starting, plugging, inching

table J5-4: utilization categories for contactors (IEC 947-4).

Types of co-ordinationFor each association of devices, a type of co-ordination is given, according to the state ofthe constituant parts following a circuitbreaker trip out on fault, or the opening of acontactor on overload.IEC 947-4-1 defines two types of co-ordination, type 1 and type 2, which setmaximum allowable limits of deterioration ofswitchgear, which must never present adanger to personnel.c type 1: deterioration of the contactor and/orof its relay is acceptable under 2 conditions:v no risk for the operator,v all elements other than the contactor and itsrelay must remain undamaged;c type 2: burning, and the risk of welding ofthe contacts of the contactor are the onlyrisks allowed.

Which type to choose?The type of co-ordination to adopt dependson the parameters of exploitation, and mustbe chosen to satisfy (optimally) the needs ofthe user and the cost of installation.c type 1:v qualified maintenance service,v volume and cost of switchgear reduced,v continuity of service not demanded, orprovided by replacement of motor-starterdrawer;c type 2:v continuity of service imperative,v no maintenance service,v specifications stipulating this type of co-ordination.

J



5.3 basic protection schemes: circuit breaker / contactor / thermal relay (continued)

key points in the successful

association of a circuit breaker

and a discontactor

1 CB magnetic-trip performance curve2 thermal-relay characteristic3 thermal-withstand limit of the thermal relay

Compact NStype MA

Isc ext.

2

13

t

I

fig. J5-5: the thermal-withstand limit of the thermal relay must be to the right of the CBmagnetic-trip characteristic.

Standards define precisely all the elementswhich must be taken into account to realize acorrect co-ordination of type 2:c absolute compatibility between the thermalrelay of the discontactor and the magnetic tripof the circuit breaker. In figure J5-5 thethermal relay is protected if its limit boundaryfor thermal withstand is placed to the right ofthe CB magnetic trip characteristic curve.In the case of a motor-control circuit breakerincorporating both magnetic and thermaldevices, co-ordination is provided in thedesign;

c the short-circuit current breaking rating ofthe contactor must be greater than theregulated threshold of the CB magnetic triprelay, since it (the contactor) must be capableof breaking a current which has a value equalto, or slightly less than, the setting of themagnetic relay (as seen from figure J5-5);c a reliable performance of the contactor andits thermal relay when passing short-circuitcurrent, i.e. no excessive deterioration ofeither device and no welding of contactorcontacts.

short-circuit current-breaking

capacity of a combination circuit

breaker + contactor

In the studies, the s.c. current-breakingcapacity which must be compared to theprospective short-circuit current is:c either, that of the CB + contactorcombination, if these devices are physicallyclose together (e.g. in the same drawer orcompartment of a MCC*). A short-circuitdownstream of the combination will be limitedto some extent by the impedances of thecontactor (see previous Note) and thethermal relay. The combination can thereforebe used on a circuit for which the prospectiveshort-circuit current level exceeds the rateds.c. current-breaking capacity of the circuitbreaker. This feature very often presents asignificant economic advantage;c or, that of the CB only, for the case wherethe contactor is separated from the CB(so that a short-circuit is possible on theintervening circuit). For such a case,IEC 947-4-1 requires the rating of the circuitbreaker to be equal to or greater than theprospective short-circuit current at its point ofinstallation.* Motor Control Centre.

it is not possible to predict the s.c.current-breaking capability of a CB +contactor combination. Laboratorytests and calculations bymanufacturers are necessary todetermine which type of CB toassociate with which contactor, andto establish the s.c. breaking capacityof the combination.Tables are published by Merlin Geringiving this information in their“LV Distribution” catalogue.

M

fig. J5-6: circuit breaker and contactormounted in juxtaposition.

M

fig. J5-7: circuit breaker and contactorseparately mounted, with interveningcircuit conductors.

J


choice of instantaneous

magnetic-trip relay for the circuit

breaker

The operating threshold must never be lessthan 12 In for this relay, in order to avoidpossible tripping due to the first current peakduring start-up. This current peak can varyfrom 8 In to 11 or 12 In.

5.4 preventive or limitative protection

preventive or limitative protectiondevices detect signs of impendingfailure, so that action can be taken(automatically or by operatorintervention) to avoid or limit theotherwise inevitable consequences.

The main protection devices of this type formotor are:c thermal sensors in the motor (windings,bearings, cooling-air ducts, etc.);c multifunction protections;c insulation-failure detection devices onrunning, or stationary motor.

thermal sensors

Thermal sensors are used to detect abnormaltemperature rise in the motor by directmeasurement.The thermal sensors are generally embeddedin the stator windings (for LV motors), thesignal being processed by an associatedcontrol device acting to trip the circuit breaker(figure J5-8).

fig. J5-8: overheating protection bythermal sensors.

multi-function motor-protection

relay

The multi-function relay, associated with anumber of sensors and indication modules,provides protection for motors, such as:c thermal overload;c rotor stalled, or starting-up period too long;c overheating;c phase current imbalance, loss of onephase, inverse rotation;c earth fault (by RCD);c running on no-load, blocked rotor onstart-up.The advantages of this relay are essentially:c a comprehensive protection, providing areliable, high-performance and permanentmonitoring/control function;c efficient surveillance of all motor-operatingschedules;c alarm and control indications;c possibility of communication viacommunication buses.

fig. J5-9: multi-function protection,typified by the Telemecanique relay, typeLT8 above.

J



5.4 preventive or limitative protection (continued)

preventive protection of

stationary motors

This protection concerns the monitoring of thelevel of insulation resistance of a stationarymotor, thereby avoiding the undesirableconsequences of insulation failure duringoperation, such as:c for motors used on emergency systems forexample: failure to start or to performcorrectly;c in manufacturing: loss of production.This type of protection is indispensable foressential-services and emergency-systemsmotors, especially when installed in humidand/or dusty locations.Such protection avoids the destruction of amotor by short-circuit to earth during start-up(one of the most frequently-occurringincidents) by giving a warning in advance thatmaintenance work is necessary to restore themotor to a satisfactory operational condition.Examples of application (figure J5-10)Fire-protection system “sprinkler” pumps.Irrigation pumps for seasonal operation, etc.Example: a vigilohm SM 20 (Merlin Gerin)relay monitors the insulation of a motor, andsignals audibly and visually any abnormalreduction of the insulation resistance level.Furthermore, this relay can prevent anyattempt to start the motor, if necessary.

SM20

MERLIN GERIN

SM20

IN OUT

fig. J5-10: preventive protection ofstationary motors.

limitative protection

Residual current differential protectivedevices (RCDs) can be very sensitive anddetect low values of leakage current whichoccur when the insulation to earth of aninstallation deteriorates (by physical damage,contamination, excessive humidity, and soon). Some versions of RCDs, speciallydesigned for such applications, provide thefollowing possibilities:c to avoid the destruction of a motor(by perforation and short-circuiting of thelaminations of the stator) caused by aneventual arcing fault to earth. This protectioncan detect incipient fault conditions byoperating at leakage currents in the range of300 mA to 30 A, according to the size of themotor (approx. sensitivity: 5 % In).Instantaneous tripping by the RCD will greatlylimit the extent of damage at the faultlocation;c to reduce considerably the risk of fire due toearth-leakage currents (sensitivity i 500 mA).A typical RCD for such duties is typeRH328A relay (Merlin Gerin) which provides:c 32 sensitivities (0.03 to 250 A);c possibility of discriminative tripping or totake account of particular operationalrequirements, by virtue of 8 possible time-delays (instantaneous to 1 s.);c automatic operation if the circuit from thecurrent transformer to the relay is broken;c protected against false operation;c insulation of d.c. circuit components:class A.

RH328A

MERLIN GERIN

fig. J5-11: example using relay RH328A.

J


voltage drop at the terminals of amotor during starting must neverexceed 10% of rated voltage Un.

The importance of limiting voltage dropat the motor during start-upIn order that a motor starts and accelerates toits normal speed in the appropriate time, thetorque of the motor must exceed the loadtorque by at least 70%. However, the startingcurrent is much greater than the full-loadcurrent of the motor; moreover, it is largelyinductive. These two factors are both veryunfavourable to the maintenance of voltageat the motor. Failure to provide sufficientvoltage will reduce the motor torquesignificantly (motor torque is proportional toU2) and will result either in an excessivelylong starting time, or, for extreme cases, infailure to start.

Example:c with 400 V maintained at the terminals of amotor, its torque would be 2.1 times that ofthe load torque;c for a voltage drop of 10% during start-up,the motor torque would be 2.1 x 0.92 = 1.7times the load torque, and the motor wouldaccelerate to its rated speed normally;c for a voltage drop of 15% during start-up,the motor torque would be 2.1 x 0.852 = 1.5times the load torque, so that the motor-starting time would be longer than normal.In general, a maximum allowable voltagedrop of 10% Un is recommended during thestart-up of a motor.

5.5 maximum rating of motors installed for consumers supplied at LV

The disturbances caused on LV distributionnetworks during the start-up of large DOL(direct-on-line) a.c. motors can occasionconsiderable nuisance to neighbouringconsumers, so that most power-supplyauthorities have strict rules intended to limitsuch disturbances to tolerable levels.The amount of disturbance created by agiven motor depends on the “strength” of thenetwork, i.e. on the short-circuit fault level atthe point concerned. The higher the faultlevel, the “stronger” the system and the lowerthe disturbance (principally volt-drop)experienced by neighbouring consumers.For distribution networks in many countries,typical values of maximum allowable starting

currents for DOL motors are shown below intable J5-12.Corresponding maximum power ratings of thesame motors are shown in table J5-13.Since, even in areas supplied by one powerauthority only, “weak” areas of the networkexist as well as “strong” areas, it is alwaysadvisable to secure the agreement of thepower supplier before acquiring the motorsfor a new project.Other (but generally more costly) alternativestarting arrangements exist, which reduce thelarge starting currents of DOL motors toacceptable levels; for example, star-deltastarters, slip-ring motors, “soft start” electronicdevices, etc.

type of motor single- location maximum starting current (A)or three-phase overhead- underground-

line network cable networksingle phase dwellings 45 45

others 100 200three phase dwellings 60 60

others 125 250

table J5-12: maximum permitted values of starting current for direct-on-line LV motors(230/400 V).

type of motor single- single-phase three-phase 400 Vor three-phase 230 Vlocation (kW) direct-on-line other methods

starting at full load of starting(kW) (kW)

dwellings 1.4 5.5 11others overhead line 3 11 22

networkunderground 5.5 22 45cable network

table J5-13: maximum permitted power ratings for LV direct-on-line-starting motors.

5.6 reactive-energy compensation (power-factor correction)

The effect of power factor correction on theamount of current supplied to a motor isindicated in table B4 in Chapter BSub-clause 3-1, and the method of correctionin Chapter E Clause 7.

J


6. protection of direct-current installations

differences between a.c. and d.c.

installations

Although the basic design principles in eachcase are similar, there are differences in:c the calculations for short-circuit currents,and;c the choice of protective equipment, sincethe techniques employed for the interruptionof direct current differ in practice from thoseused for alternating current.

6.1 short-circuit currents

in order to calculate the maximumshort-circuit current from a battery ofstorage cells, when the internalresistance of the battery is unknown,the following approximate formulamay be used:Isc = kC where C = the ratedampere-hour capacity of the battery,and k is a coefficent close to 10 (andin any case is less than 20).

battery of storage cells (or

accumulators)

For a short-circuit at its output terminals, abattery will pass a current according to Ohm’slaw equal to Isc = Vb/Riwhere: Vb = open circuit voltage of the fully-charged batteryRi = the internal resistance of the battery (thisvalue is normally obtained from themanufacturer of the battery, as a function ofits ampere-hour capacity)When Ri is not known, an approximateformula may be used, namely: Isc = kCwhere C is the ampere-hour rating of thebattery and k is a coefficient close to 10, andin any case is always less than 20.

Example:What is the short-circuit current level at theterminals of a battery with the followingcharacteristics:c 500 Ah capacity;c fully-charged open-circuit voltage 240 V(110 cells at 2.2 V/cell);c discharge rate 300 A;c autonomy 1/2 hour;c internal resistance is 0.5 milli-ohm/cellso that Ri = 110 x 0.5 = 55 mΩ for the battery,

and Isc = 240 x 103 = 4.4 kA 55The short-circuit currents are seen to be(relatively) low.

Isc

fig. J6-1: battery of storage cells.

direct-current generator

If Vg is the open-circuit voltage of thegenerator and Ri its internal resistance, then:Isc = Vg / Ri.In the absence of precise data, and for a d.c.system of voltage Un, Vg may be taken as1.1 Un.

Example:A d.c. generator rated at 200 kW, 230 V, andhaving an internal resistance of 0.032 ohm,will give a terminal short-circuit current of230 x 1.1 = 7.9 kA 0.032

IccG=

fig. J6-2: direct-current generator.

Isc at any point in an installation

In this case Isc = V Ri + RlWhere Ri is as previously defined,V is either Vb or Vg as previously defined,Rl is the sum of the resistances of the fault-current loop conductors.Where motors are included in the system,they will each (initially) contribute a current ofapproximately 6 In (i.e. six times the nominalfull-load current of the motor) so that:

Isc = V + 6 (In mot) Ri + Rlwhere In mot is the sum of the full-loadcurrents of all running motors at the instant ofshort-circuit.

+

-

fig. J6-3: short-circuit at any point of aninstallation.

J


6.2 characteristics of faults due to insulation failure, and of protective switchgear

Devices for circuit interruption are sensitive tothe level of d.c. voltage at their terminalswhen breaking short-circuit currents.The table below provides the means fordetermining these voltages, which depend onthe source voltage and on the method ofearthing the source.

Note: In the following text the word “pole” hastwo meanings, viz:1. Referring to a d.c. source, for example:the positive pole or the negative pole of abattery or generator.2. Referring to a switch or circuit breaker, forexample:a pole of a circuit breaker makes or breaksthe current in one conductor. A pole of acircuit breaker may be made up of modules,each of which contains a contact. The polemay therefore consist of one module or(particularly in d.c. circuits) several series-connected modules.

Voltage stresses across opening contactsare reduced by the technique of connecting anumber of contacts in series per pole, asmentioned in the table below, and in thefollowing text.

types of network system earthing unearthed systemone pole earthed source with source is notat the source mid-point earthing earthed

earthing schemesand various faultconditions

case 1 case 2 case 3analysis fault A pole (a) must break pole (a) must break there is no short-circuitof each maximum Isc at U volts maximum Isc* in this casefault at U/2 volts

fault B poles (a) and (b) must poles (a) and (b) must poles (a) and (b)break the maximum break the maximum must break the maximum IscIsc at U volts Isc at U volts at U volts

fault C there is no short-circuit as for fault A but as for fault Acurrent in this case concerning pole (b)*

* U/2 divided by Ri/2 = Isc (max.)

the most fault A A = B = C fault B (or faults A and Cunfavourable case see Note below the table simultaneously)case of a circuit breaker all the contacts participating in provide in the CB pole provide the number of contacts

current interruption are series for each conductor the number necessary for breaking theconnected in the positive- of contacts necessary to break current indicated in the CB poleconductor (or the negative Isc (max.) at the voltage U/2. of each conductor.conductor if the positive poleof the source is earthed). Providean additional pole for inserting in theearthed polarity conductor, to permitcircuit isolation (figure J6-6).

+

–U

B A

C

i

R

a

b

+

–

U/2+

U/2 B A

C

ia

b

R+

–U

B A

C

i

R

a

b

table J6-4: characteristics of protective switchgear according to type of d.c. system earthing.

Note: each pole is equally stressed for faults at A, B or C, since maximum Isc must be broken with U/2 across the CB pole(s) in each case.

6.3 choice of protective device

for each type of possible insulationfailure, the protective devices againstshort-circuits must be adequatelyrated for the voltage levels noted intable J6-4 above.

The choice of protective device depends on:c the voltage appearing across the current-breaking element. In the case of circuitbreakers, this voltage dictates the number ofcircuit-breaking contacts that must beconnected in series for each pole of a circuitbreaker, to attain the levels indicated intable J6-4;c the rated current required;c the short-circuit current level at its point ofinstallation (in order to specify its s.c. current-breaking capacity);

c the time constant of the fault current (L/Rin milli-seconds) at the point of installation ofthe CB.Table J6-5 below gives characteristics(current ratings, s.c. current-breakingcapacity, and the number of series-connectedcontacts per pole required for a given systemvoltage) for circuit breakers made byMerlin Gerin.

J


6. protection of direct-current installations (continued)

6.3 choice of protective device (continued)

type ratings sc current-breaking capacity kA thermal coefficient for(A) for L/R i 0.015 seconds overload uprating the

(the number of series-connected protection instantaneouscontacts per pole is shown in brackets) magnetic24/48 V 125 V 250 V 500 V 750 V 1000 V tripping units*

C32HDC 1 to 40 20 (1p) 10 (1p) 20 (2p) 10 (2p) special DC special DCC60a 10 to 40 10 (1p) 10 (2p) 20 (3p) 25 (4p) ditto AC 1.38C60N 6 to 63 15 (1p) 20 (2p) 30 (3p) 40 (4p) ditto AC 1.38C60H 1 to 63 20 (1p) 25 (2p) 40 (3p) 50 (4p) ditto AC 1.38C60L 1 to 63 25 (1p) 30 (2p) 50 (3p) 60 (4p) ditto AC 1.38NC100H 50 to 100 20 (1p) 30 (2p) 40 (3p) 20 (4p) ditto AC 1.42NC100LH 10 to 63 50 (1p) 50 (1p) 50 (1 p) 50 (3p) ditto AC 1.42NS100N 16 to 100 50 (1p) 50 (1p) 50 (1p) 50 (2p) ditto AC 1.42NC100H 16 to 100 85 (1p) 85 (1p) 85 (1p) 85 (2p) ditto AC 1.42NS100L 16 to 100 100 (1p) 100 (1p) 100 (1p) 100 (2p) ditto ACNS160N 40 to 160 50 (1p) 50 (1p) 50 (1p) 50 (2p) ditto ACNS160H 40 to 160 85 (1p) 85 (1p) 85 (1p) 85 (2p) ditto ACNS160L 40 to 160 100 (1p) 100 (1p) 100 (1p) 100 (2p) ditto ACNS250N 40 to 250 50 (1p) 50 (1p) 50 (1p) 50 (2p) ditto ACNS250H 40 to 250 85 (1p) 85 (1p) 85 (1p) 85 (2p) ditto ACNS250L 40 to 250 100 (1p) 100 (1p) 100 (1p) 100 (2p) ditto ACNS400H MP1/MP2-400 85 (1p) 85 (1p) 85 (1p) 85 (2p) no thermal relay; tripping unitsNS630H MP1/MP2/MP3-630 85 (1p) 85 (1p) 85 (1p) 85 (2p) provide MP1/MP2/MP3C1251N-DC P21/P41-1250 50 (1p) 50 (1p) 50 (2p) 50 (3p) 25 (3p) an external special forM10-DC 1000 100 (3p) 100 (3p) 100 (3p) 100 (3p) 50 (4p) 50 (4p) relay direct currentM20-DC 2000 100 (3p) 100 (3p) 100 (3p) 100 (3p) 50 (4p- 50 (4p) (if necessary)M40-DC 4000 100 (3p) 100 (3p) 100 (3p) 100 (3p) 50 (4p) 50 (4p)M60-DC 6000 100 (4p) 100 (4p) 100 (4p)M80-DC 8000 100 (4p) 100 (4p) 100 (4p)

table J6-5: choice of d.c. circuit breakers manufactured by Merlin Gerin.

* These tripping units may be used on a.c. or d.c. circuit breakers, but the operating levels marked on each unit correspond tor.m.s. a.c. values. When used on a d.c. circuit breaker the setting must be changed according to the co-efficient in table J6-5.For example, if it is required that the d.c. circuit breaker should trip at 800 A or more the coefficient given in table J6-5 is 1.42,then the setting required will be 800 x 1.42 = 1,136 A.

6.4 examplesExample 1Choice of protection for an 80 A outgoing d.c.circuit of a 125 V system, of which thenegative pole is earthed. The Isc = 15 kA.

NC100 H3-pole80 A

load

125 V =+

-

fig. J6-6: example.

Table J6-4 shows that the full system voltagewill appear across the contacts of the positivepole.Table J6-5 indicates that circuit breakerNC100H (30 kA 2 contacts/pole 125 V) is anappropriate choice.Preferred practice is to (also) include acontact in the negative conductor of theoutgoing circuit, to provide isolation (formaintenance work on the load circuit forexample), as shown in figure J6-6.

Note: three contacts in series, which open inunison, effectively triple the speed of contactseparation. This technique is often necessaryfor successfully breaking d.c. current.

Example 2Choice of protection for a 100 A outgoing d.c.circuit of a 250 V system, of which the mid-point is earthed. Isc = 15 kA.

NC100 H4-pole100 A

load

250 V =+

-

fig. J6-7: example.

Table J6-4 shows that each pole will besubject to a recovery voltage of U/2, i.e.125 V for all types of s.c. fault.Table J6-5 indicates that circuit breakerNC100H (30 kA 2 contacts/pole 125 V) issuitable for cases A and C, i.e. 2 contacts inthe positive and 2 contacts in the negativepole of the CB.It will be seen in the 250 V column that4 contacts will break 20 kA at that voltage(case B of table J6-4).

J


6.5 protection of persons

The rules for protection are the same asthose already covered for a.c. systems.However, the conventional voltage limits andthe automatic disconnection times for safetyof persons are different (see tables G8 andG9 of Chapter G, Sub-clause 3.1):c all exposed conductive parts areinterconnected and earthed;c automatic tripping is achieved in the time-period specified.RCDs are not applicable to d.c. circuits, sothat in practice:c the principles of the TN scheme are usedfor cases 1 and 2 of Sub-clause 6.2. It is thensufficient to check that, in the case of a short-circuit, the current magnitudes will besufficient to trip the instantaneous magneticrelays.The checking methods are identical to thoserecommended for an a.c. network.c principles of the IT scheme for case 3 inSub-clause 6.2,v the insulation level of the installation mustbe under permanent surveillance and anyfailure must be immediately indicated andalarmed: this can be achieved by theinstallation of a suitable monitoring relay asshown in Chapter G, Sub-clause 3.4,v the presence of two concurrent faults toearth (one on each polarity) constitutes ashort-circuit, which will be cleared byovercurrent protection. As for the a.c.systems, it is sufficient to verify that thecurrent magnitude exceeds that necessary tooperate the magnetic (or short-time delay)circuit breaker tripping units.

+

-

TR5A

+

-

XM200

U fixed U variable ou fixed

fig. J6-8: insulation (to earth) monitors foran IT direct-current installation.

domestic and similar premises and special locations

Domestic and similar premisesand special locations

1. Domestic and similar premises1.1 general L11.2 distribution board components L21.3 protection of persons L41.4 circuits L61.5 protection against overvoltages and lightning L8

2. Bathrooms and showers2.1 classification of zones L102.2 equipotential bonding L122.3 requirements prescribed for each zone L12

3. Recommendations applicable L13

to special installations and locations

L

domestic and similar premises and special locations - L1

L

1. domestic and similar premises

the power distribution authorityconnects the LV neutral pointof its HV/LV distribution transformerto earth.All LV installations must be protectedtherefore by RCDs (for TT and ITschemes of earthing) or by short-circuit protective devices for TNschemes.All exposed conductive parts must bebonded together and connected toearth, either directly to an electrodeat the premises (TT or IT schemes)or by means of the neutral conductor(TN schemes)*.

Electrical installations for inhabited premisesneed a high standard of safety and reliability.

1.1 general

related standardsMost countries have national regulations and-or standards governing the rules to be strictlyobserved in the design and realization ofelectrical installations for domestic and similarpremises. The relevant international standardis the IEC publication 364.

the power networkThe vast majority of power-distributionauthorities connect the low-voltage neutralpoint of their HV/LV distribution transformersto earth.The protection of persons against electricshock therefore depends, in such cases, onthe principles discussed in Chapter F,Clause 4, and Chapter G, all Clauses.The measures required depend on whetherthe TT, TN or IT scheme of earthing isadopted, as treated in detail in Chapter G.RCDs are essential for TT- and IT-earthedinstallations, but high-speed overcurrentdevices (MCBs or fuses) are commonly usedto clear earth faults on TN-earthed schemes.However, in particular instances (e.g. circuitsfeeding socket-outlets), RCDs are stronglyrecommended on TN installations, as beingthe only sure means of protection againstshock when very long flexible leads of smallc.s.a. are supplied from a socket. See alsoClause 3 concerning special installations.* for TN-C and TN-S schemes refer to Chapter G,Sub-clause 5.1.

L2 - domestic and similar premises and special locations

L

1. domestic and similar premises (continued)

1.2 distribution-board components

overcurrentprotection and isolation

protection against direct and indirect contact,and protectionagainst fire

enclosure

serviceconnection

lightningprotection

remote control

energymanagement

control anddistribution board

incoming-supplycircuit breaker

lightning arrester

cartridgefuses

MCB phase + neutral

differentialMCB

Isis

differential load switch

bistable switch

THP

IHPc

CDSc

fig. L1-1: presentation of realizable functions on a consumer unit.


L

the quality of electrical equipmentused in inhabited premises iscommonly ensured by a mark ofconformity situated on the front ofeach item.

Distribution boards (generally one only indomestic premises) usually include themeter(s) and in some cases (notably wherethe supply authorities impose a TT-earthingsystem and/or tariff conditions which limit themaximum permitted current consumption) anincoming-supply differential circuit breaker,which includes an overcurrent trip. This circuitbreaker is freely accessible to the consumer.On installations which are TN-earthed, thesupply authorities usually protect theinstallation simply by means of sealed fusecut-outs immediately upstream of themeter(s). See figure L1-2. The consumer hasno access to these fuses.

meter

fuses or circuit breakerdepending on earthing system

tableau de répartition

fig. L1-2: components of a control anddistribution board.

the incoming-supplycircuit breakerThe consumer is permitted to operate this CBif necessary (e.g. to reclose it if the currentconsumption had exceeded the authorizedlimit; to open it in case of emergency or forisolation purposes...). The differential tripgenerally has a 500 mA setting to provideindirect-contact protection (and a measure offire protection) for the whole installation.Current ratings for these circuit breakers arecommonly:c 15 - 90 A two-poles,c 10 - 60 A four-poles.

fig. L1-3: incoming-supply circuit breaker.

the control and distributionboard (consumer unit)

This board comprises:c a control panel for mounting (whereappropriate) the incoming-supply circuitbreaker and other control auxiliaries, asrequired,c a distribution panel for housing 1, 2 or3 rows (of 24 multi 9 units) or similar MCBs orfuse units, etc.,c installation accessories for fixingconductors, and rails for mounting MCBs,fuse bases, etc. neutral busbar and earthingbar, and so on...,c service-cable ducts or conduits, surfacemounted or in cable chases embedded in thewall.

Note: to facilitate future modifications to theinstallation, it is recommended to keep allrelevant documents (photos, diagrams,characteristics, etc.) in a suitable locationclose to the distribution board.The board should be installed at a heightsuch that the operating handles, indicatingdials (of meters) etc. are between 1 metreand 1.80 metres from the floor (1.30 metresin situations where handicapped persons orpersons of advanced age are concerned).

fig. L1-4: control and distribution board.

lightning arresters

Where the keraunic level of a locality exceeds25, and the supply is taken from an overheadline, the installation of a lightning arrester atthe service position of a LV installation isprescribed in many national standards and isstrongly recommended for installations whichinclude sensitive (e.g. electronic) equipment.These devices must automatically disconnectthemselves from the installation in case offailure or be protected by a RCD of

appropriate sensitivity, according to theresistance of the earthing electrode for theinstallation. In the case of domesticinstallations the use of a 500 mA differentialincoming-supply circuit breaker type S (i.e.slightly time-delayed) will provide effectiveearth-leakage protection, while, at the sametime, will not trip unnecessarily each time alightning arrester discharges the current (ofan overvoltage-surge) to earth.


L


if, in TT schemes, the resistanceof the installation earth electrodeexceeds 100 ohms, then oneor more 30 mA RCDs must beinstalled to take over the functionof the differential device in theincoming-supply circuit breaker.

1.2 distribution-board components (continued)

resistance value of the earthelectrode (of a TT scheme)

In the case where the resistance-to-earth isvery high and exceeds the value

Rt = 50 V = 100 ohms 500 mAone or several RCDs of appropriatesensitivity, i.e. 30 mA, should be used inplace of the differential device of theincoming-supply circuit breaker.

1.3. protection of persons

where public power-supply systemsand consumers' installations form aTT-earthed scheme, the governingstandards impose the use of RCDsto ensure the protection of persons.

On TT-earthed systems the protection ofpersons is ensured by the followingmeasures:c protection against indirect-contact hazardsby RCDs of medium sensitivity (300 or500 mA) at the origin of the installation(incorporated in the incoming-supply circuitbreaker, or on the incoming feed to thedistribution board). This measure isassociated with a consumer-installed earthelectrode to which must be connected theprotective-earth (PE) conductors from theexposed conductive parts of all class Iinsulated appliances and equipment, as wellas those from the earthing pins of all socketoutlets.

c where the CB at the origin of an installationhas no RCD protection (see figure L1-7) theprotection of persons shall be ensured byclass II level of insulation on all circuitsupstream of the first RCDs. In the case wherethe distribution board is metallic, care shall betaken that all live parts are double insulated(supplementary clearances or insulation, useof covers, etc.) and wiring reliably fixed,c obligatory protection by sensitive (30 mA)RCDs of socket-outlet circuits, and circuitsfeeding bathrooms, laundry rooms, and so on(for details of this latter obligation, refer to thetable in Clause 3 of this Chapter).

incoming-supply circuit breakerwith instantaneous differentialrelay

In this case:c an insulation fault to earth could result inthe shutdown of the entire installation,c where a lightning arrester is installed, itsoperation (i.e. discharging a voltage surge toearth) could appear to an RCD as an earthfault, with a consequent shutdown of theinstallation.Recommendation of suitableMerlin Gerin componentsc incoming-supply circuit breaker with500 mA differential, andc RCD of type DDR-HS 30 mA (for example,differential circuit breaker 1P + N type DéclicVigi) on the circuits supplying socket-outlets,c RCD of type DDR-HS 30 mA (for example,differential load switch type ID’clic) on circuitsto bathrooms, shower rooms, laundry rooms,etc.) (lighting, heating, socket-outlets).

500 mA

30 mA 30 mA

diversecircuits

socket-outletscircuit

bathroom and/orshower room

DB

fig. L1-5: installation with incoming-supply circuit breaker havinginstantaneous differential protection.


Lincoming-supply circuit breakertype S with retarded differentialrelay

This type of CB affords protection againstinsulation faults to earth, but by virtue of ashort time delay, provides a measure ofdiscrimination with downstreaminstantaneous RCDs. Tripping of theincoming-supply CB and its consequences(on deep freezers, for example) is therebymade less probable in the event of lightning,or other causes of voltage surges. Thedischarge of voltage-surge current to earth,through the lightning arrester, will leave thetype S circuit breaker unaffected.Recommendation of suitableMerlin Gerin componentsc incoming-supply circuit breaker with500 mA differential, type S, andc RCD of type DDR-HS 30 mA (for example,differential circuit breaker 1P + N type Déclic-Vigi) on the circuits supplying socket-outlets,c RCD of type DDR-HS 30 mA (for example,differential load switch, type ID’clic) on circuitsto bathrooms, shower rooms, etc. (lighting,heating, socket-outlets),c RCD of type DDR-HS 30 mA (for example,differential circuit breaker 1P + N, type Déclic-Vigi) on circuits supplying washing machinesand dish-washing machines.

500 mA

30 mA 30 mA

diversecircuits

high-risk location(laundry room)


DB type S

30 mA

bathroom and/orshower room

fig. L1-6: installation with incoming-supply circuit breaker having short timedelay differential protection, type S.

incoming-supply circuit breakerwithout differential protection

In this case the protection of persons must beensured by:c class II level of insulation up to thedownstream terminals of the RCDs,c all outgoing circuits from the distributionboard must be protected by 30 mA or 300 mARCDs according to the type of circuitconcerned as discussed in Chapter G,Clause 4,c where a voltage-surge arrester is installedupstream of the distribution board (to protectsensitive electronic equipment such asmicroprocessors, video-cassette recorders,TV sets, electronic cash registers, etc.) it isimperative that the device automaticallydisconnects itself from the installationfollowing a rare (but always possible) failure.Some devices employ replaceable fusingelements; the recommended methodhowever as shown in figure L1-7, is to use aRCD.

Recommendation of suitableMerlin Gerin componentsFigure L1-7 refers.1. Incoming-supply circuit breaker withoutdifferential protection.2. Automatic disconnection device (if alightning arrester is installed).3. RCD of type DDR-HS 30 mA (for example,differential circuit breaker 1P + N type DéclicVigi) on each circuit supplying one or moresocket-outlets.4. RCD of type DDR-HS 30 mA (for example,differential load switch type ID’clic) on circuitsto bathrooms and shower rooms (lighting,heating and socket-outlets) or a 30 mAdifferential circuit breaker per circuit.5. RCD of type DDR-HS 300 mA (forexample, differential load switch) on all theother circuits.

6. Safety and tripping discrimination areimproved by the protection of circuits bymeans of 30 mA RCDs (for example,differential circuit breaker type Déclic-Vigi1P + N) on a circuit supplying an apparatuswhich involves large quantities of water.

300 mA 30 mA 30 mA 30 mA

diversecircuits

high-risk circuit(dish-washingmachine)


bathroomand/orshower room

1 2

365 4

class IIinsulation

fig. L1-7: installation with incoming-supply circuit breaker having nodifferential protection.


L


1.4 circuits

the distribution and division of circuitsprovides comfort, and facilitates rapidlocation of faults.

subdivisionNational standards commonly recommendthe sub-division of circuits according to thenumber of utilization categories in theinstallation concerned (see figure L1-8):c at least 1 circuit for lighting. Each circuitsupplying a maximum of 8 lighting points,c at least 1 circuit for socket-outlets rated10/16 A. Each circuit supplying a maximum of8 sockets. The sockets may be single ordouble units (a double unit is made up of two10/16 A sockets mounted on a common basein an embedded box, identical to that of asingle unit),c 1 circuit for each appliance such as a waterheater, washing machine, dish-washingmachine, cooker, refrigerator, etc.Recommended numbers of 10/16 A (orsimilar) socket-outlets and fixed lightingpoints, according to the use for which thevarious rooms of a dwelling are intended, areindicated in the following table.

socket-outlets

lighting heating washingmachine

cookingapparatus

fig. L1-8: circuit division according toutilization.

room function minimum number of fixed minimum numberlighting points of 10/16 A socket-outlet

living room 1 5bedroom, lounge, 1 3bureau, dining roomkitchen 2 4 (1)bathroom, shower room 2 1 or 2entrance hall, box room 1 1WC, storage space 1 -laundry room - 1

table L1-9: recommended minimum number of lighting and power points in domesticpremises.(1) of which 2 above the working surface and 1 for a specialized circuit: in addition an independent socket-outlet of 16 A or 20 Afor a cooker and a junction box or socket-outlet for a 32 A specialized circuit.


L

the inclusion of a protectiveconductor in all circuits is required byIEC and most national standards.

protective conductors

IEC and most national standards require thateach circuit includes a protective conductor.This practice is strongly recommended whereclass I insulated appliances and equipmentare installed, which is the general case.The protective conductors must connect theearthing-pin contact in each socket-outlet,and the earthing terminals in class Iequipment, to the main earthing terminal atthe origin of the installation.Furthermore, 10/16 A (or similarly sized)socket-outlets must be provided withshuttered contact orifices.^

cross-sectional-area (c.s.a.)of conductorsThe c.s.a. of conductors and the rated currentof the associated protective device dependon the current magnitude of the circuit, theambient temperature, the kind of installation,and the influence of neighbouring circuits(refer to Chapter H1).Moreover, the conductors for the phasewires, the neutral and the protectiveconductors of a given circuit must all be ofequal c.s.a. (assuming the same material forthe conductors concerned, i.e. all copper orall aluminium).Table L1-11 indicates the c.s.a. required forcommonly-used appliances.Protective devices 1 phase + N in 2 x 9 mmspaces comply with requirements forisolation, and for marking of circuit currentrating and conductor sizes. fig. L1-10: circuit breaker 1 phase + N

2 x 9 mm spaces Déclic 32.

type of circuit c.s.a. maximum protectivesingle-phase 230 V of the power device1 ph + N or 1 ph + N + E conductorsfixed lighting 1.5 mm2 2300 W circuit breaker 16 A

(2.5 mm2) fuse 10 A

10/16 A 2.5 mm2 4600 W circuit breaker 25 Asocket-outlets (4 mm2) fuse 20 A

individual-load circuitswater heater 2.5 mm2 4600 W circuit breaker 25 A

(4 mm2) fuse 20 A

dish-washing 2.5 mm2 4600 W circuit breaker 25 Amachine (4 mm2) fuse 20 A

clothes-washing 2.5 mm2 4600W circuit breaker 25 Amachine (4 mm2) fuse 20 A

cooker or 6 mm2 7300 W circuit breaker 40 Ahot plates (1) (10 mm2) fuse 32 A

electric 1.5 mm2 2300 W circuit breaker 16 Aspace heater (2.5 mm2) fuse 10 A

table L1-11: c.s.a. of conductors and current rating of the protective devices in domesticinstallations (the c.s.a. of aluminium conductors are shown in brackets).(1) in a 230/400 V 3-phase circuit, the c.s.a. is 4 mm2 for copper or 6 mm2 for aluminium, and protection is provided by a 32 Acircuit breaker or by 25 A fuses.


L


1.5 protection against overvoltages and lightning

the relevance of protective

devices

c disturbancesThree types of disturbance often occur onelectrical-power networks:v lightning and atmospheric electricalphenomena in general, with its direct andindirect consequences.The direct effects, which are fairly infrequent,concern its impact on overhead transmissionand distribution lines.The indirect effects are more common andoccur at lower energy levels. Such indirectphenomena are characterized by a powerfulinduction effect on the lines and/or by anincrease of local earth potential,v operational overvoltages are transient, andare caused by abrupt changes in the circuit,such as the opening/closing of circuitbreakers, load-break switches, contactors,etc.,v overvoltages at normal system frequencycan occur in many ways, and will do so, forexample, if a neutral connection is broken ona 3-phase system, if the load is unbalanced,c the kind of installation to be protected.It is necessary to know in some detail thecharacteristics of the items to be protected, inorder to select the most appropriate form ofprotection. The choice of protective device(s)depends on two factors:v the sensitivity: the ability of the equipmentconcerned to withstand an overvoltagecondition i.e. its magnitude and duration,v the cost: which comprises the purchaseprice and the operational costs (possiblelosses, maintenance, etc.).

choice of a lightning arresterIt depends on:c the level of the disturbance,c the cost, as noted above,c the connection to a LV electrical-powernetwork, a telephone system, a building-control system bus, or to any other network,c the kind of installation-earthing scheme(see chapter F) (figure L1-12).


Linstallation rules

Three principal rules must be respected:c it is imperative that the three lengths ofcable used for the installation of the arrester(see figure L1-13) each be less than 50 cm,i.e.:v the live conductors connected to theisolating switch,v from the isolating switch to the lightningarrester,v from the lightning arrester to the maindistribution board (MDB) earth bar (not to beconfused with the main protective-earth (PE)conductor or the main earth terminal for theinstallation).The MDB earth bar must evidently be locatedin the same cabinet as the lightning arrester,c it is necessary to use an isolating switch ofa type recommended by the manufacturer ofthe lightning arresterc in the interests of a good continuity ofsupply it is recommended that the circuitbreaker be of the time-delayed or selectivetype.The guide "Lightning protection" treats theserules quantitively and includes informationwhich allows the user to dimension anarrester according to his needs.


L

0.60 m

zone 2

zone 1

zone 0

zone 3

0.60 m 2.40 m

zone 2

zone 1

zone 0

zone 3

2.40 m

zone 2zone 1

zone 0

2.25 m

zone 3

zone 1

zone 1 0.60 m 2.40 m

* *

plan views

vertical cross-section

2. bathrooms and showers

Bathrooms and shower rooms are areas ofhigh risk, because of the very low resistanceof the human body when wet or immersed inwater.Precautions to be taken are thereforecorrespondingly rigorous, and the regulationsare more severe than those for most otherlocations.The relevant IEC standards are 364-7-701,479 and 669-1.Precautions to observe are based on threeaspects:c the definition of zones, numbered 0, 1, 2and 3 in which the placement (or exclusion)of any electrical device is strictly limited orforbidden, and, where permitted, the electricaland mechanical protection is prescribed,c the establishment of an equipotential bondbetween all exposed and extraneous metalparts in the zones concerned,c the strict adherence to the requirementsprescribed for each particular zone, as tabledin Clause L3.

2.1 classification of zones

Sub-clause 701.32 of IEC 364-7-701 definesthe zones 0, 1, 2 and 3, as shown in thefollowing diagrams:

* zone 1 is above the bath as shown in the vertical cross-section.

fig. L2-1: zones 0, 1, 2, 3 in proximity to a bath-tub.


L

0.60 m

zone 0zone 1 zone 2 zone 3

2.40 m

0.60 m

zone 0zone 1 zone 2 zone 3

2.40 m

zone 2zone 1

zone 0

2.25 m

zone 3

zone 1

0.60 m 2.40 m

fig. L2-2: zones 0, 1, 2, 3 in proximity of a shower with basin.

zone 1 zone 3

0.60 m

permanent wall

2.25 m

fixed shower head (1)

zone 1

zone 2

0.60 m

0.60 m

2.40 mzone 3

zone 1

zone 2

0.60 m

2.40 m

zone 3

0.60 m

fixed shower head (1)

zone2

fig. L2-3: zones 0, 1, 2, 3 in proximity of a shower without basin.(1) When the shower head is at the end of a flexible tube, the vertical central axis of a zone passes through the fixed end of the flexible tube.

0.60 m

prefabricatedshowercabinet

0.60 m

fig. L2-4: no switch or socket-outlet ispermitted within 60 cm of the dooropening of a shower cabinet.


L

2. bathrooms and showers (continued)

2.1 classification of zones (continued)

classesof externalinfluences

zone 3

dressing cubicles (zone 2)

AD 3BB 2BC 3

AD 3BB 3BC 3AD 7BB 3BC 3

WC

AD 3

BB 2

BC 3

AD 3BB 2BC 3

shower cabinets (zone 1)


fig. L2-5: individual showers with dressing cubicles.


zone 2BB 3BC 3

AD 7

BB 3

BC 3WC

AD 3

BB 2

BC 3

h<1.10 mAD 51.10 m<h<2.25 mAD 3

dressing cubicles

BB 3BC 3

h<1.10 mAD 51.10 m<h<2.25 mAD 3

zone 1


fig. L2-6: individual showers with separate individual dressing cubicles.

classes of externalinfluences

AD 3

BB 2

BC 3

AD 3

BB 2

BC 3

h<1.10 mAD 51.10 m<h<2.25 mAD 3

BB 3BC 3

dressing roomzone 3

zone 1zone 2

AD 7

BB 3

BC 3

classes of externalinfluences

fig. L2-7: communal showers and a common dressing room.

2.2 equipotential bonding

equipotential conductorsfor a bathroom

metallicpipesh i 2m

metaldoor-frame

radiator

metal bath

lighting

gas

water-drainagepiping

to the earthelectrode

fig. L2-8: supplementary equipotential bonding in a bathroom.

2.3 requirements prescribed for each zone

The table of Clause 3 describes theapplication of the principles mentioned in the

foregoing text and in other similar or relatedcases.


L


and locations

The following table summarizes the mainrequirements prescribed in many nationaland international standards.

Note:Section numbers in brackets refer to sectionsof IEC 364-7.

locations protection principles IP level wiring switchgear socket- installationand cables outlets materials

domestic c TT or TN-S schemes 20 1 switch operating protectiondwellings c differential protection handles and by 30 mAand other v 500 mA if the earth electrode similar devices RCDshabitations resistance is i 100 ohms on distribution

instantaneous or short time delay panels, to be(type S) mountedv 30 mA if the earth electrode betweenresistance is u 500 ohms 1 metre andc lighning arrester at the origin of the 1.80 metreinstallation if: above the floorv supply is from overhead line with levelbare conductors, and if:v the keraunic level > 25c a protective earth (PE) conductor onall circuits

bathrooms supplementary equipotentialor shower rooms bonding in(section 701) zones 0, 1, 2 and 3

zone 0 SELV 12 V only 27 1 X X X specialzone 1 SELV 12 V 24 1 (1) class II limited to X (2) X special

strict minimum and water heaterzone 2 SELV 12 V or 30 mA RCD 23 1 (1) X (2) one socket- class II insulation

outlet only with and protection by aa separation 30 mA RCDtransformer

zone 3 21 1 (1) class II protectionc by 30 mA RCDc by electrical separation, orc by SELV 50 V (Chapter G, Sub-clause 3.5)

swimming baths supplementary equipotential bonding special (under-water(section 702) in zones 0, 1 and 2 lighting)

zone 0 SELV 12 V 28 1 X X X special (under-waterlighting)

zone 1 SELV 12 V 25 1 X X X Xzone 2 interior 22 1 (1) class II protection

exterior 24 1 (1) c by 30 mA RCDc by electrical separation, orc by SELV 50 V (Chapter G, Sub-clause 3.5)

saunas 31 1 class II X X adapted to(section 703) temperaturework sites c conventional voltage 35 7 mechanically protection(section 704) limit UL reduced to 25 V protected by 30 mA

c TT or TN-S scheme RCDsagricultural and c conventional voltage 35 5 protectionhorticultural limit UL reduced to 25 V by 30 mAestablishments c protection against fire risks RCDs(section 705) by 500 mA RCDsrestricted 31 1 protection of:conductive c portable tools:working spaces v by SELV or(metal tank v by electricalinteriors, etc) separation(section 706) c hand-held lamps

v by SELVc fixed equipmentv by SELVv by electricalseparationv by 30 mA RCDsv by specialsupplementaryequipotentialbonding

(1) if subjected to water jets: IP x 5(2) except a switch for SELV


L


and locations (continued)

locations protection principles IP level wiring switchgear socket- installationand cables outlets materials

fountains protection by 30 mA RCDs and(section 702) equipotential bonding of all exposed

and extraneous conductive partsdata processing TN-S scheme recommended (3)(section 707) TT scheme if leakage current is

limited. Protective conductor 10 mm2

minimum in aluminium. Smaller sizes(in copper) must be doubled

caravan parks 34 5 flexible under- protection of(section 708) ground cables circuits by

30 mA RCDs(one per 6socket-outlets)

marinas and 36 5 in conduits dittoyacht basins or buriedmedical centres IT (medical) scheme equipotential protection by

bonding 30 mA RCDsfairs and protection by 30 mA 21 7exhibitions RCDs. TT or TN-S schemebalneotherapy individual: see section 701(cure-centre (volumes 0 and 1)baths) collective: see section 702

(volumes 0 and 1)motor-fuel filling explosion risks in security zones limited to thestations necessary

minimummotor vehicles protection by RCDs or by electrical

separation

(3) in an IT scheme provide a LV/LV transformer to create a TN-S scheme.

appendix

Appendix - EMC

Appendix Chapter CExample in coordination of the characteristics of an HV switch-fuse C1 - 1combination protecting an HV/LV transformerGround-surface potential gradients due to earth-fault currents C2 - 1Vector diagram of ferro-resonance at 50 Hz (or 60 Hz) C3 - 1

Chapter EElementary harmonic filters E3 - 1Harmonic suppression reactor for a single capacitor bank E4 - 1(power factor correction)

Chapter JShort-circuit characteristics of an alternator J1 - 1

EMC 1. Rules and regulations EMC - 1

2. Electromagnetic disturbances2.1 disturbances by conduction EMC - 22.2 radiation EMC - 8

3. Cabling of equipment and systems3.1 earthing EMC - 93.2 masses EMC - 113.3 attenuation effects EMC - 153.4 installation and cabling rules EMC - 163.5 EMC components and solutions EMC - 17

4. Local network problems EMC - 20

List "cachiers techniques" 21

Ap

Appendix C1 - 1

C

example in coordination of the characteristicsof an HV switch-fuse combination protectingan HV/LV transformer

This appendix is based on Appendices A andB of IEC 420, and is intended to clarify someof the operational features of thesecombination units.The following example is based on a400 kVA, 11 kV/LV transformer with amaximum fault level at its HV terminals of16 kA.c the full-load current is 21 A,c the permissible periodic overload is 150%F.L.,c the off-circuit tapping switch is selected tothe - 5% tap, so that the primary current inthe overload condition is:21 x 1.5 x 1.05 = 33 A,c the magnetizing in-rush current surge is:21 x 12 = 252 A maximum for a duration of0.1 second (Clause 4a of IEC 787),c the ambient-air temperature at site is 45 °C,i.e. 5 °C higher than the IEC standard.The user has selected a 12 kV switch-fusecombination from a certain manufacturer toprotect the transformer. The manufacturer willprovide a list of fuses which are suitable foruse in the combination, and will recommendthose fuses necessary for this particularapplication.The choice of the manufacturer will be basedon factory type-tests to the appropriate IECspecifications covering this class of HVswitch-fuse combination.Suppose the manufacturer recommends a12 kV, 40 A, 16 kA (at least) fuse of a giventype from a certain fuse manufacturer. Tojustify this choice, the manufacturer will haveascertained that:1. The fuse can withstand the 252 A of inrushcurrent for 0.1 second, without anymodification to its subsequent performance.This will be done by reference to the time-current characteristic of the fuse, and/or inconsultation with the fuse manufacturer.2. The normal current rating of thecombination when using the recommendedfuses is adequate to carry 33 A periodically inan ambient-air temperature of 45 °C, i.e. itmatches the transformer overload capability.Note: the normal current rating of thecombination when fitted with therecommended 40 A fuses may, in fact, beless than 40 A, particularly in the above-standard ambient temperature conditions.Temperature-rise tests carried out by theswitch-fuse manufacturer, or calculationsbased on such tests, may indicate a normalcurrent rating of (for example) 35 A at 45 °C.Such a rating is evidently satisfactory for thisapplication.

3. The pre-arcing current in the fuse is lowenough in the 10-second region of the fusetime-current characteristic, to ensuresatisfactory protection of the transformer(Clause 4c of IEC 787). This information willbe obtained by the switch-fuse manufacturer,from the fuse characteristic curves, and/or inconsultation with the fuse manufacturer.4. The fuses alone will clear a solid 3-phaseshort-circuit fault at the LV terminals of thetransformer, i.e. the maximum short-circuitprimary currrent (based on 5% transformerreactance) is greater than the transfer current(current at which the switch operatesconcurrently with the fuse(s))* when thecombination includes the recommended 40 Afuses.Figure AC1-1 shows that the transfer currentin this case is 280 A.5. The transfer current of the combinationwhen fitted with the 40 A fuses concerned isless than its rated transfer current, assumedin this case to be 1,000 A.* transfer current is defined below.

The installation designers must check that thefuse discriminates with the highest rating of aLV fuse (if existing) in the event of a phase-to-phase fault on the LV system. This isgenerally the worst condition fordiscrimination, since the LV fuses pass0.87 Isc3 while one HV fuse (only) passesIsc3, as shown in figure AC1-2 (b). Toensure discrimination in this case, the time-current characteristics of the HV and LVfuses should intersect at a current valuewhich is greater than that of the maximumpossible short-circuit current on the LVsystem, as shown in figure AC1-3.Discrimination between HV fuses and LVcircuit breakers has been covered inChapter C Sub-clause 3.2 (figure C21) andChapter H2 Sub-clause 4.6 (figures H2-56and H2-57).

2 - Appendix C1

C

transfer current and take-over current

transfer currentThe transfer current of a combinationdepends on both the fuse-initiated (striker)opening time of the switch, and the time-current characteristic of the fuse.Near the transfer-current level, during aLV 3-phase short-circuit (at the transformerterminals), the fastest fuse to melt clears onepole and operates its striker pin.The two remaining poles are then passing areduced current (87%), which will beinterrupted by the switch or by the fuses.The transfer point is that at which the switchopens and one or both remaining fuses meltsimultaneously.For this to occur, the second fuse must meltat the instant of switch opening (by strikeraction of the first fuse to operate).By calculation (demonstrated in Appendix Bof IEC 420) it is shown that the level of3-phase fault current* which will cause thesecond fuse to melt at a time (equal to theswitch-opening time) after the operation ofthe first fuse, is that corresponding to a periodof 0.045 seconds after fault initiation, asshown on the time-current fuse-characteristiccurve.* this is the transfer-current value. Its reduction to 87%,following operation of the first fuse, is taken into accountwhen calculating the operating time of the second fuse.

take-over currentThe take-over current of the combination isthe level of overcurrent at which the fusestake over the duty of protection from overloadrelays, i.e. from the take-over current level upto and beyond the transfer current level, theswitch will be tripped by striker-pin action.

fig. AC1-1: principles of HV/LV transformer protection by HV switch-fuse combination.

Figure AC1-1 is intended only to show thebasic principles involved, and takes noaccount of maximum and minimumtolerances in the fuse pre-arcing curves, etc.For greater detail, reference should be madeto IEC 420.

fig. AC1-2: short-circuit currents for thetransformer of the example, assuming a242-420 V secondary winding.

(a) 3 - phase S.C.

(b) phase-to-phase S.C.

(c) 1 - phase-to-earth S.C.

430 A

430 A

430 A

10.7 kA

10.7 kA

10.7 kA

430 A

215 A

215 A

9.3 kA

0

248 A

248 A

10.7 kA

time (secs.)

fault current (amps)

overload cleared by switch only on operation of overload relay

faults clearedby fuses and switch through striker action

faults cleared by fuses only(strikers operate but fault is cleared before switch contacts open)

10 21 33 100 252 420 1000

minimum ratedbreaking current of fuse

take-over current level

0.01

0.1

1.0

10

rated transfer currentof the combination

transfer current level

overload relay characteristic

40 A fuse characteristic

full-loadcurrent

50%overloadcurrent

0.045 secs.

not to scale

280

Appendix C1 - 3

C

types of faults involved in the transfer region

Primary-side protective devices areparticularly concerned with faults in thesecondary-terminals zone of the transformer,upstream of the LV protection devices.The primary short-circuit currents arising fromsolid short-circuits at the transformersecondary terminals are shown infigure AC1-2.The breaking of solid (i.e. non-arcing)3-phase faults is associated with severe TRVvalues which the switch in the combination isnot designed to interrupt. This type of faulttherefore must be cleared by the fuses only,i.e. before the striker-operated switch opensits contacts.It is necessary therefore, that the transfer-current limit (280 A in the example) shallalways be lower than that of a 3-phaseLV-terminals short-circuit (430 A at HV), asshown in figure AC1-1.This condition being fulfilled, then 3-phaseshort-circuit transfer currents correspond tofaults for which LV arc impedance reducesthe magnitude of both the current and TRVvalues, as well as improving the power factorof the fault current.For a phase-to-phase LV terminal fault, thediagram (b) of figure AC1-2 shows that theHV fault current of one phase is equal to thatof a 3-phase LV short-circuit. The fuse of thatphase will clear rapidly; the current in the tworemaining phases will then reduce topractically zero*, and be cleared in thetransfer-current region by two of the switchcontacts acting in series. This featureimproves the switch performance for breakingcurrent which is (in the present case) mainlytransformer-magnetizing current.

For a phase-to-earth LV terminal fault,diagram (c) of figure AC1-2 shows that theHV fault current is less than the calculatedvalue (280 A) for the transfer-current limit.As in the previous case, following theoperation of one fuse (assuming that bothfuses do not clear simultaneously), the faultcurrent will reduce to a very low value. Theswitch contacts breaking this low-valued (buthighly-inductive) current, are acting in series,with the advantage of enhanced switchgearperformance noted above.

fig. AC1-3: discrimination between HV andLV fuses.* because the LV voltages induced in the faulted phases willthen be sensibly equal in magnitude, but will be in directphase opposition around the fault-current loop.

HV fault current (amps)430 A

maximum faultcurrent at LV(referred to HV side)

minimum time-current characteristic of HV fuse

maximum operatingtime of LV fuse(referred to HV side)

time (seconds)

Appendix C2 - 1

C

ground-surface potential gradients due to earth-faultcurrents

When earth-fault current flows between anearthing electrode and the surrounding soil,potential gradients exist in the soil and onthe ground surface.Close to the location of the buried electrode,the potential gradients, both in the soil andon the ground surface, are generally at theirmaximum values and are therefore (for theground-surface gradients) the mostdangerous.

nature of the potential gradientsA vertical-rod electrode is very commonlyused, either singly, or in interconnectedgroups; a single-rod electrode is taken as anexample in the following notes. The currentflow and associated potential contoursassociated with a rod electrode are shown infigure AC2-1, and are based on the followingsimplifying assumptions:c perfectly homogeneous soil,c the origin of the fault current (i.e. a short-circuit to earth) is at an infinite distance fromthe electrode, thereby creating a perfectlysymmetrical current-flow between theelectrode and the surrounding soil.Earth rod at 1.0 pu (per unit) volts withrespect to remote-earth.

"remote-earth" and "local-earth"

zero-potential references

In classical theory "remote-earth" is an earth-reference point at zero potential, located atan infinite distance from the electrodeconcerned. The following notes show that,where a current circulates between twoelectrodes, a local zero-potential-earthreference is available, which can be used inpractical electrode-resistance tests. Theresulting measured values correspondclosely to those calculated by the classicaltheory.

fig. AC2-1: idealized current-flow patternand associated equipotential contours ofa single-rod vertically-driven earthelectrode.Note: current-flow lines are identical to the lines of theelectric field.

Vg = voltage of test-instrument generatorRI = resistance in ohms indicated by an electrode-resistance measuring instrumentVI = voltage applied to test instrumentV = voltage at electrode A with respect to local-earth referenceVp = voltage at probe P with respect to local-earth referenceFor an accurate measurement of resistance of electrode A, test-electrode P must be located at 0, the position of which is notknown exactly (see text for fig. F52 Chapter F).

fig. AC2-2: zero-voltage local-earth reference for two electrodes.

0.4

0.50.60.7

0.9 0.8

equipotential contours

current-flowlines

ground surface

+ 0.2+ 0.3

"local-earth" zero-voltage equipotential reference plane

typicalcurrent-flowlines

positivepotential region

negativepotential region

electrode A

+ 1.0

Azero

B

vp (c) voltage profile not to scale

-1

B

Vg

(d) measured resistance for different locations of electrode P

distance from Aof test-electrode P

OA

VIRI

zero

VI = (V + vp)in zone B

VI = (V - vp)in zone A

V

electrode B

(b) equipotential surfaces in the earth

+ 1.0 pu -1.0 puA B0

(a) plan view of ground-surface equipotential contours

0

+ 0.4+ 0.5+ 0.6

+ 0.1- 0.2- 0.3- 0.4- 0.5- 0.6

- 0.1



+0.2+0.3 +0.1

+0.4+0.5 -0.5-0.4

-0.3-0.2-0.1

ground-surface

vp

When two electrodes exchange fault current,i.e. the power-source electrode and theelectrode of an installation at which an earthfault occurs, there exists (at some pointbetween the two electrodes) an equipotentialvertical-plane surface of infinite area,perpendicular to the flow lines of the faultcurrent, as shown in figure AC2-2.

2 - Appendix C2

CThe vertical plane surface is the locus atwhich the strength of the positive* electricfield from one electrode (A) is exactly equal tothe strength of the negative field of the otherelectrode (B). This means that the resultantpolarity at the plane is neither positive nornegative. It therefore constitutes anequipotential surface at zero volts withrespect to the two electrodes.When "local-earth" is mentioned in thesenotes, it refers to a point on the ground atwhich the edge of this plane surface islocated.From the foregoing description, it can beseen that the zero-voltage equipotentialsurface is also the boundary of the two"zones of influence" of the two electrodes,sometimes also referred to as "zones ofresistance".As a matter of interest, the electric fields ofthe two electrodes (X) and (C) for theelectrode-resistance test, shown in figure F52(Chapter F) are essentially similar to thosementioned above. The location of local-earthin each of the diagrams of F52 (and ofdiagram (d) in AC2-2) is indicated by 0 (i.e.zero).* Since a.c. systems are being considered, each electrodewill change its polarity at every half cycle.

potential gradients of avertical-rod elec trodeIt will be seen in figure AC2-1 that the highcurrent density at the (pointed) tip of the rodresults in steep potential gradients in the soilbeneath the buried extremity of the electrode,as indicated by the close proximity ofadjacent equipotential contours in that region.It can also be seen that the gradient at theground surface is less severe than thosebelow the surface, and that the maximumgradient at the surface occurs immediatelyadjacent to the point at which the rodemerges from the soil, as shown infigure AC2-3.

fig. AC2-3: voltage profile of the single-rodelectrode of figure AC2-1.Note: the equipotential contours on the surface of theground, when viewed from above, are concentric circlesaround the rod location (in the idealized conditionsassumed).

Investigations have shown that, in ahomogeneous soil, between 0.5 pu and0.8 pu of the voltage at the rod is measuredbetween the rod and a point on the ground at1 metre from it (i.e. approximately the lengthof a step) for very long-rod and very short-rodelectrodes, respectively.It is clear therefore, that, where earth-faultcurrents are high, the area close to such anelectrode will be dangerous. Measures takento reduce such dangers are described later inthis appendix.

reason for potential gradientsThe resistance to a flow of current in aconducting medium is given in ohms by the

formula R = ρl where aρ is the resistivity coefficient of the medium,expressed in ohm-metres,l is the length in metres of the conductingpath (in the direction of the current-flow lines),a is the cross-sectional area in square metresthrough which the current flows.So that, for a given distance (l) along, the

current-flow lines R ∝ 1 a

The soil in contact with the rod has an area(a) equal to that of the rod surface, while atpoints progressively further from the rod, thecurrent passes through increasingly-largeareas.For equal lengths of current-flow path,therefore, the resistance becomesprogressively lower and lower, and thevoltage drops across equally-spaced intervalsbecome less and less, so that the voltagegradients are smaller and smaller withdistance from the rod, as shown in figureAC2-3. Another way of considering theformation of potential gradients is shown infigure AC2-1, in which the equipotentialcontours are shown at 0.1 pu voltageintervals. Since the same current is passingthrough all the equipotential surfaces, theresistance of the mass of soil between anytwo adjacent equipotential surfaces will havethe same value.This means that, as the areas of successivesurfaces are larger, the lengths of the currentpaths between successive surfaces mustincrease for R to remain constant (reminderR = ρl/a). The spacing between successivesurfaces must therefore be greater (i.e. l/amust be constant) and this results inprogressively lower potential gradients.

0.8 msteplength

Vpu

rodposition

maximum gradient at the rod-soil interface

0.5 1 metres1 0.5

0.50.32 pustep voltage

1.0 pugroundsurface

local-earth datum (zero volts)

Appendix C2 - 3

Cvoltage gradients associated withearthing gridsThe purpose of an earthing grid (or mat) is toprovide a close approximation to anequipotential condition at the ground surfaceover a large area, generally that of aswitchyard or substation.In practice, potential gradients will alwaysoccur when earth-fault currents are flowing,but providing the grid meshes areappropriately dimensioned (i.e. not too large),the permissible maximum values of gradientat the highest anticipated levels of earth-faultcurrent will not be exceeded.During an earth fault, the whole of theearthing grid and all metallic parts connectedto it (together with any personal present) maybe raised to several hundreds (or thousands)of volts.Potential gradients in the grid meshes willhave the general form shown in figure AC2-4.This figure also shows that connecting ametallic boundary fence to the earthing gridcan be dangerous, unless adequateprecautions are taken.

reducing potential gradients dueto earth faultsSome of the methods commonly employed inthe reduction of potential gradients include:c reduction of earth-fault-current levels:v by using resistance- or reactance-earthedsources (generators or transformers asappropriate),v where transformers or generators(according to the case) operate permanentlyin parallel, some of them are left unearthed,c for electrodes generally (figure AC2-5) by:(a) increasing the length and/or number ofrods to reduce the electrode resistance andtherefore the voltage rise at the electrode,(b) using special low-resistance "soils" in thespace surrounding the electrodes,(c) reducing earthing-grid-mesh sizes, andusing "grading electrodes" at the gridboundaries. The same "grading-electrode"technique is sometimes used around thebase of transmission-line towers, boundaryfences, etc. to reduce the severity of thegradients, and therefore that of both "touch"and "step" voltages,(d) insulating driven rods from contact withthe earth over the upper section, i.e. from thesurface of the ground to a depth ofapproximately 1 m.

fig. AC2-4: voltage profile and potential gradients of an earthing grid.

The resistance of a rod electrode is approximately inverselyproportional to its length.

S1, S2 and S3 are strip grading electrodes running parallelto the fence and connected to it at frequent intervals.

Note: the connection at the top of the rod, as well as theconnecting lead, must also be insulated. An alternativemethod is to bury the rod completely with the top of the rodbelow ground level, and connected to an insulatedconnecting lead.

fig. AC2-5: voltage profiles and methodsof reducing maximum potential gradientsin some common earthing arrangements.

touchvoltage

grid voltage withrespect to local-earth

ground surface

fence

step voltage

zero potential

β

αlong rod

(a)

short rod

fence is out of reach

grid voltagewith respectto local-earth

voltage profiles atthe ground surface

metal fence

βα

S1S2

S3

(c)

with grading

withoutgrading

zero

β

α

insulation

(d)

withoutinsulation

with insulation

voltage profiles

β

α

gradient β withspecial earth fill

speciallow-resistance earth fill

(b)

gradient α withoriginal soil

4 - Appendix C2

Cother m ethods of reducingthe dangers of ground-surfacepotential gradientsThe easiest method (but wasteful in terms ofspace) is simply to fence off the area aroundthe electrode(s) with warning notices. Morecommonly, measures are taken to reduce thecurrent passing through a person's feet byproviding an insulated floor covering indoors,such as plastic tiles, rubber mats, etc., whilefor outdoor locations, highly-resistivesurfaces, such as crushed rock, thick layersof asphalt, or clean gravel or pebbles arefrequently used.Gravel or pebbles provide a very effectivehigh-resistance surface, even when wet,providing the stones are clean. Leaf mould, ormud etc. between the stones greatly reducesthe insulating performance of such surfaces.

safe levels of ground-surfacepotential gradientThere is no IEC-recommended safe value ofmaximum long-duration (> 10 seconds) touchvoltage for HV installations at the time ofwriting (1994), but many authorities haveadopted the 50 V AC (or 25 V AC in wetconditions) criteria of IEC 364-4-41.Standards differ in various countries, and incertain cases, long-duration touch-voltagevalues which are significantly higher than50 V AC maximum are permitted, while thereis an even greater difference above therecommended IEC 364 values for allowableshort-duration (e.g. 0.03 to 0.5 seconds)touch voltages.In the present circumstances, therefore, it isnecessary to comply with the appropriatelocal regulations.At the time of writing, Cenelec TechnicalCommittee 112 is preparing a European HVinstallation standard. Chapter 9 (according topresent proposals) of this future standard willinclude recommendations for safe touch-voltage/time duration levels.It is recognized that, for a given voltage level,a touch-voltage is more dangerous to humanbeings than a step-voltage (since, in theformer case, a substantial part of the currentpasses through the vital organs in the thorax).If the gradients are such that the touch-voltage criterion is satisfied, then the step-voltage condition is also considered to besatisfactory.

Appendix C3 - 1

C

vector diagram of ferro-resonance at 50 Hz (or 60 Hz)

Figure C13 of Sub-clause 3.1 Chapter Cshows how the neutral of an unearthed3-phase source can be displaced from itsnormal near-zero potential, as a result of theparticular ferro-resonant conditiondescribed.The vector diagram can be constructed asfollows.Per-unit notation is used to generalize andsimplify the calculations:c 1 pu voltage is the nominal phase-to-neutral system voltage,c 1 pu impedance is equal to the normalcapacitive reactance of one phase-to-earthat power frequency (i.e. K in figure C14 ofChapter C),c the source impedance is assumed to benegligible.

fig. AC3-1: calculation of VNE

- circuit andvector diagrams.

The procedure is as follows:c compute the current IN in an imaginaryneutral of negligible impedance (i.e. ZN = 0)by summating the individual phase currents,as shown in figure AC3-1.According to Thévenin's theorem IN alsoequals VNE whereZ

NE + Z

VNE is the voltage between N and E when noneutral connection exists,ZNE is the impedance of the networkmeasured between terminals N and E whenno neutral connection exists,ZN is the impedance of the neutral conductor(zero in the present case).This means that VNE = IN ZNE

calculation of IN

In figure C14 of Chapter C it was stated that,in the resonant condition, X

L > X

C. In these

calculations XC = -j 1 pu impedance, and

XL = j 10 pu.

I1 = 1 90° = 0.1 0° = 0.1 + j 0 10 90°

I2 = 1 -30° = 0.1 -120° = -0.05 - j 0.0866 10 90°

I3 = 1 210° = 1 300° = 0.5 - j 0.866

1 -90°IN = 0.55 - j 0.953 = 1.1 -60°

calculation of Z NE

ZNE is the parallel combination of XL1, XL2 andXc.

XL1 in parallel with XL2 = j 5j 5 in parallel with X

C3 =

j 5 x (-j 1) = 5 = -j 1.25 j 5 - j 1 j 4Z

NE = - j 1.25 pu ohms.

calculation of VNE

1.1 -60° x 1.25 -90° = 1.375 -150°or 1.375 pu 210°

complete vector diagramOther values shown in the vector diagram areeasily obtained from the above calculations.➀, ≠ and ➂ are the power-supply terminals.

fig. AC3-2: vector diagram for theresonant condition.

V1

I1

I2 V2

V3

VNE

I1 + I2

I1 + I2 + I3 = IN

C3L2

L1

Eimaginary neutral

23

N

1

source

N

not to scale

2

IL2IL1 + IL2 = 0.375 pu

1

N11

IL1

IC3

32.066 pu

2.066 pu

not to scale

VNE = 1.375 pu

V = 0.375 pu3 E

E

Appendix E3 - 1

E

elementary harmonic filters

For this appendix, the more commonly-occurring odd-numbered harmonics areshown in the diagrams.Before the advent of power-electronics,even-numbered harmonics were rarelyencountered, so that the 100 Hz (on 50 Hzsystems) separating one harmonicfrequency from the next made the task offilters (despite manufacturing tolerances,and impedance changes with temperature,etc.) relatively straightforward, so thatsatisfactory results were achieved (and stillare in all but exceptional cases) by themethods described below.If it is required to eliminate (almost) aharmonic voltage existing across two pointsA and B in a network, a series-connectedLCR circuit (figure AE3-1(a)) tuned toresonate at the harmonic frequencyconcerned, will constitute a virtual short-circuit to the current of that harmonicfrequency, thereby reducing VAB(h) topractically zero.The same procedure can be adopted for anynumber harmonic frequencies known to bepresent, the individual filters being connectedin parallel across the points A-B(figure AE3-1(b)).

fig. AE3-1.

An exact analysis of the combination is notsimple, since each filter is affected by thosein parallel with it, as well as by the power-system source reactance shunting the filterbank (shown dotted in figure AE3-3).When all factors have been taken intoaccount, including a degree of dampingcaused by the load impedance, theresponse of the filter bank in terms of itsimpedance at different frequencies is shownin figure AE3-2.

fig. AE3-2.

It will be seen that, at each harmonicfrequency for which a filter has beenprovided, the impedance is very low, while atintermediate frequencies, high-impedancevalues occur.Care should be taken to ensure thatfrequencies corresponding to the low-impedance point are not close to controlfrequencies (such as those of ripple-controlschemes used by many power companiesfor remote control of power-network devices).Otherwise, the control signals will be virtuallyshort-circuited.Harmonic-producing equipment must createthe harmonic e.m.f.s. and resulting currentsin order to function correctly. The role of afilter bank, as described, is to allow a freeflow of harmonic currents to circulatebetween the harmonic source and the filterbank, while practically eliminating thesecurrents and voltages from the rest of thenetwork.

B

harmonicsource

(b) A

protectednetwork andpower source

L

C

R

(a) A

B

f =2 π v LC

1

IZI Ω

fh/f50

(harmonicorder)

75 11 131

2 - Appendix E3

E

fig. AE3-3.

In figure AE3-3, it will be seen that since thefilters are practically short-circuits toharmonics, most of the harmonic voltage V

hwill be dropped across the internalimpedance Z

h of the harmonic source and

that small harmonic-current components onlywill pass through the power-system sourceimpedance X

s and the loads (the latter having

relatively high impedance).Since at fundamental frequency thecapacitive reactance of each filter is muchgreater than its inductive reactance, most ofthe power-frequency voltage will appearacross the capacitors, so that a usefulcontribution to any power-factor correctionrequirement is fortuitously available.

damped harmonic filtersAs noted in Chapter E, Sub-clause 9.1, themagnitude of harmonic emfs diminishes asthe order of the harmonic increases. Thefiltering requirements are not, therefore, socritical for high-order harmonics as thosenecessary for lower-order harmonics.For that reason, the filter for the highestharmonic of a bank, such as that shown infigure AE3-1 (b) is often damped, byconnecting a resistor in parallel with thereactor.The result is a filter which is less effective(but adequate) at its tuned frequency, whileat all higher frequencies, the impedance willbe low (inductive/resistive), approaching thevalue of the resistor only (figure AE3-4) asthe frequency increases (i.e. it forms a "high-pass" filter).Such a high-pass filter is commonly used forthe highest-order harmonic filter (the 11th or13th for example) of a bank, as showndotted in figure AE3-3.

fig. AE3-4: damped filter circuit andcharacteristic impedance/frequencycurve.

There are several variations of dampedfilters and many combinations of band-passand undamped filters in service, accordingto particular requirements.In fact, the successful application of powerelectronics devices is largely due to thedevelopment of effective filtering techniqueswhich are, however, beyond the scope ofthese brief notes.

B

harmonicsource

A

X sourcea

13111175

filter bank

loadsVh

Zh

L

Cr

R

IZI Ω

Hz

fo

R

Appendix E4 - 1

E

(b)

IZI Ω

f (Hz)

f1 fp fS

parallelresonancefP

power sourceimpedance

range ofunwantedharmonic

frequencies

L

C

R

LS(a)

seriesresonance fS

harmonic suppression reactor for a single(power factor correction) capacitor bank

As shown in Appendix E2, the crux of theproblem for capacitor banks is that a fractionof the total component of a given harmoniccurrent can be magnified to dangerous levelsin a parallel LCR circuit if that circuitresonates at the harmonic frequencyconcerned.By connecting a reactor L in series with thecapacitor bank, the parallel resonantcondition is moved away from the harmonicfrequency towards a lower frequency, asshown in figure AE4-1 (b).In fact, the circuit now resonates at twodifferent frequencies; the lower frequency isdue to the parallel Ls//LCR combination, andthe upper due to the series LCR circuit.

fig. AE4-1.

It is sufficient that the two resonantfrequencies be lower than those of theharmonics to be protected against, to ensurecomplete immunity from resonance. Thereason for this is that for frequencies higherthan the series-resonant frequency XL > XC sothat the LCR branch behaves as aninductance + resistance series circuit. Thisbranch being in parallel with LS, the power-system source inductance, no resonantcondition is possible.Furthermore, the addition of reactor L meansthat changes in the power-system sourcereactance will have much less influence (thanformerly) on the parallel-resonant frequency,since L generally has a much greater valuethan LS (e.g. 2 to 9 times) and the parallel-resonant frequency depends on L + LS.It may be noted that, although a harmonic-suppression reactor protects the capacitorbank against the problem of resonance withthe source reactance, it does not reduce theamount of harmonic current which passesthrough the HV/LV transformer to the source.Such currents must be eliminated by shunt-connected series filters, as described inAppendix E3.

stepped banks of capacitorsPower-factor correction capacitor banks arefrequently made up of a number of switchedsections, so that the amount of compensationcan be adjusted to suit the requirements of achanging load.If all switched steps have the same kvarrating, then the series resonant frequency ofeach step must be the same, i.e. the first stepin service must fulfill the conditions of seriesresonance already mentioned and shown infigure AE4-1 (b). The addition of furtheridentical steps in parallel will not affect thetwo resonant frequencies fp and fs.This is because, although the capacitancehas increased n times (for n steps in service),the inductance has reduced to 1/n times itsoriginal value, so that the product LC, onwhich the series resonant frequencydepends, remains constant.By similar reasoning, it follows that mixedsteps of any kvar rating may be paralleled,providing that every step is tuned to the sameseries resonant frequency.

Appendix J1 - 1

J

short-circuit characteristics of an alternator

The characteristics of a 3-phase alternatorunder short-circuit conditions are obtainedfrom oscillogram traces recorded duringtests, in which a short-circuit is appliedinstantaneously to all three terminals of amachine at no load, excited (at a fixed level)to produce nominal rated voltage.The resulting currents in all three phases willnormally* include a d.c. component, whichreduces exponentially to zero after(commonly) some tens of cycles. The curveshown below in figure AJ1-1 is the currenttrace, from which the d.c. component hasbeen eliminated, of a recording made duringthe testing of a 3-phase 230 V 50 kVAmachine.The definitions of alternator reactancevalues are based on such "symmetrical"curves.

fig. AJ1-1: short-circuit current of onephase of a 3-phase alternator with the d.c.component eliminated.* unless, by chance, the voltage of a phase happens to bemaximum at the instant of short-circuit. In that case, therewill be no d.c. transient in the phase concerned.

The reduction of current magnitude from itsinitial value occurs in the following way.At the instant of short-circuit, the onlyimpedance limiting the magnitude of currentis principally** the inherent leakagereactance of the armature (i.e. stator)windings, generally of the order of10%-15%.The large stator currents are (practically)entirely inductive, so that the synchronouslyrotating m.m.f. produced by them acts indirect opposition to that of the excitationcurrent in the rotor winding.The result is that the rotor flux starts toreduce, thereby reducing the e.m.f.generated in the stator windings, andconsequently reducing the magnitude of thefault current. The effect is cumulative, andthe reduced fault current, in turn, nowreduces the rotor flux at a slower rate, andso on, i.e. the flux follows the exponentiallaw of natural decay, its reduction rate at anyinstant depending on the magnitude of thequantity causing the phenomenon.Eventually, a stable state is reached, in whichthe (greatly reduced) rotor flux produces justenough voltage to maintain the stator currentat the level of equilibrium between the threequantities, viz. current, flux and voltage.The reduction of fault current therefore iscaused by a diminution of the generatede.m.f. due to armature reaction, and not, infact, by an increase in impedance of themachine (that is why the term "effectivereactance" was used in Chapter JSub-clause 1.1).** the sub-transient reactance, which is defined later, isvery nearly equal to the leakage reactance.

As shown in figure AJ1-1, the currentreduction requires a certain time, and thereason for this is that, as the rotor flux beginsto diminish, the change of flux induces acurrent in the closed rotor circuit in thedirection which, in effect, increases theexcitation current, i.e. opposes theestablishment of a reduced level of magneticflux. The gradual predominance of the statorm.m.f. depends on the overall effect of rotorand stator time constants, the result of whichis the principal factor in the "a.c. currentdecrement" shown in figure AJ1-1.If, during a short-circuit, there were no eddycurrents induced in the unlaminated face ofround-rotor alternators, or in damperwindings (see note 1) of salient-polealternators, the envelope of the a.c. currentdecrement would be similar to that ofcurve b in figure AJ1-1, i.e. the so-calledtransient-current envelope.The presence of either of the two features,mentioned above however, gives rise to thesub-transient component of current (curve C).The effect is analogous to that of the closedcircuit of the rotor-excitation windingdescribed above (i.e. the induced currentsoppose the change), but having a very muchshorter time constant.The overall a.c. current decrement istherefore composed of the sum of twoexponentially-decaying quantities, viz. thesub-transient and the transient components,as shown in figure AJ1-2.Note 1: Damper windings are made up of heavy gaugecopper bars embedded in the pole faces of salient-polerotors, to form a squirrel-cage "winding" similar to that of aninduction motor. Their purpose is to help to maintainsynchronous stability of the alternator.With the rotor turning at the same speed as that of them.m.f. due to the stator currents, no currents will beinduced in the damper windings; if a difference in the speedof rotation occurs, due to loss of synchronism, then currentsinduced in the damper windings will be in a direction thatproduces a torque which acts to slow (an overspeedingrotor) or to accelerate (an underspeeding rotor). A similar,but much smaller effect occurs due to eddy currents in thesurface of solid unlaminated rotors of turbo-alternators.

For advanced analytical studies ofalternators, two component axes "direct" and"quadrature" are defined, and subtransientand transient reactances, etc. are derived foreach component system.In the simple studies needed for 3-phasesymmetrical fault levels and for circuit-breaker performance based on such faults,the direct-axis component system only isrequired; this accounts for the suffix "d" ofreactance values, shown in figure AJ1-2.Suffix "q" is used for quadrature quantities.

x''d = the sub-transient reactance Vo/i''x'd = the transient reactance Vo/i 'xd = the synchronous reactance Vo/iVo = peak rated voltage of the alternator

fig. AJ1-2: a.c. component of armaturecurrent versus time, in a short-circuitedalternator (no d.c. transient is shown).

t

c

b

i

a0

t

ia

i

transientperiod

i'

i"

steadystate

subs-transientperiod

enveloppeof the current, ia

Vo/x"d

Vo/x'd

Vo/xd

2 - Appendix J1

J

time

statorphasecurrent

d.c. component

instant ofshort circuit

The reactances are generally defined asr.m.s. voltages divided by r.m.s. currents. Inthe current trace of figure AJ1-2, however, itis simpler to use the projected peak valuesof current, so that Vo must be the rated peakvoltage of the machine.Note 2: in the definition of "i" some authors use the actualvoltage measured during the test, instead of Vo. Moreover,xd is generally denoted by Xs and is referred to as"synchronous reactance".

asymmetrical currentsAs previously noted, in general, all 3-phasesof short-circuit current will include a d.c.component. These components give rise toadditional electro-dynamic and thermalstresses in the machine itself, and in circuit-breakers protecting a faulted circuit.The worst condition is that of a phase inwhich the d.c. component is the maximumpossible, i.e. the d.c. transient value at zerotime (the instant of fault) is equal to the peakvalue of current given by Vo/xd'', as definedin figure AJ1-2.A typical test trace of this condition is shownin figure AJ1-3.

The current envelope of an asymmetrical transient has thesame dimensions about the d.c. transient curve, as thesymmetrical envelope has about the current zero axis.

fig. AJ1-3: a fully-offset asymmetricaltransient fault-current trace.The consequence of asymmetrical transientfault currents and the standardizedrelationship between the symmetrical andasymmetrical quantities for circuit breakerperformance ratings are given inSub-clause 1.1 of Chapter C, and areillustrated in figure C5.

Appendix EMC - 1

EMC

1. rules and regulations

Legislation on EMC throughout the world isbroadly divided into two philosophies. In"liberal" countries, any parasitic interferencewith radio reception is illegal, but noemission-level limit for the source ofinterference is imposed. However, in cases oflitigation the methods of measuring, and thelimits of emission level laid down by theCISPR, serve as reference. For example,Japan is a "liberal country" in which the VCCIstandards (that correspond technically to theinternational publications of CISPR) assumesthat the attitude of civic responsibilityprevailing is adequate at the present time.For countries more rigidly "regulated",emission levels exceeding a standardizedlimit are illegal. For example, in the U.S.A.,Automatic Data Processing (ADP) systemsare protected by obligatory emission-levelstandards defined by the Federal law FCCpart 15. Checking procedures differ,depending on whether the level for class A(procedure for "compliance") or class B isconcerned. In the case for class B (domesticenvironment), certification is then required.European regulations are effectively placedbetween the two foregoing attitudes. Theparasitic-emission level and an excessively-sensitive reception device are both illegal.Compliance with the EMC standards,although only constituting a presumption ofconformity to the essential requirements, is,nevertheless, the preferred means ofchecking. Moreover, the Europeanregulations are applicable to all apparatuses,systems and commercialised installations,without exception.

2 - Appendix EMC

EMC

2. electromagnetic disturbances

Problems of EMC (ElectroMagneticCompatibility) often arise when equipment,which is highly sensitive to extraneouselectrical disturbances (commonly referred toas "interference" or "parasites") is located inan environment subject to electromagneticdisturbances.Since sources of electromagnetic disturbanceare numerous and inevitable, anddesensitizing an equipment (generallyelectronic) to counter the effect ofdisturbances is difficult to achieve,consideration of the physical layout of thesensitive equipments and related cabling,relative to the sources of disturbance,becomes necessary.This is the principal means of ensuring asatisfactory degree of immunity for the largemajority of sensitive electronic devices.There are two recognized modes ofelectromagnetic interference:c conducted disturbances propagated alongcables, wires, etc.c radiated disturbances by stationaryinduction (magnetic or electrostatic fields*)and/or by electromagnetic (radio) waves

travelling in space.The magnitudes of electromagneticdisturbances are expressed by fourparameters: two for the conduction mode andtwo for the radiation mode.For the conduction mode, measurements aremade in the traditional quantities, viz: volts(U) and amps (I). For the radiatedelectromagnetic waves the electric andmagnetic field strengths are measured involts per metre (E) and in amperes permetre (H), respectively.The frequency is one of the principal featuresthat characterise an electromagnetic wave.In EMC studies the solutions adapted differaccording to whether the disturbance is at lowfrequency (LF) or at high frequency (HF).

2.1 disturbances by conduction

2.1.1 Disturbances by conduction: modes

of propagation

Electrical energy, whether useful signals orpower, or unwanted parasites, propagatealong a 2-wire circuit in one of two modesonly, viz: in differential mode or in commonmode.

Differential mode

The differential mode is the normal way ofconducting current through a 2-wire circuit.This mode is sometimes referred to as seriesmode, normal mode, or symmetrical mode.In the differential mode the current flowing inone conductor is in exact phase opposition tothat in the other conductor, i.e. flowing in theopposite direction at every instant.The voltage is measured between the twoconductors.

Common mode

The common mode is parasitic. It issometimes also referred to as parallel mode,longitudinal mode or asymmetrical mode.Common-mode currents pass through all theconductors of a cable in the same direction.The return path for such currents is via theearth, earth-bonding connections and theprotective earthing conductors, cablesheaths, etc. Since the earth is no longerused as a conducting medium for telefax,useful signals are no longer transmitted in thecommon mode through the associatedcables. A potential difference in commonmode is measured between the mass (localzero voltage reference terminal) and themean value of potential of all the conductorsof the cable circuit being tested. It may bepresent in the absence of any current flow.

fig. EMC-1: signal or disturbance in the

differential mode.

fig. EMC-2: disturbance in common

mode.

Differential mode disturbances are the mostsevere at low frequencies. By lowfrequencies (LF) it is understood in EMCstudies to concern all frequencies lower than9 kHz. This convention means that a verylarge number of electrical disturbances areconsidered as being LF phenomena.In electrical power networks disturbances inthe differential mode are numerous. One maycite, for example, interruptions of supply ofshort or long duration, voltage fluctuationsand dips, phase instability, lamp flicker,variations of frequency: harmonics andvoltage spikes. The effect of an electromagneticdisturbance depends largely on its duration.Permanent (maintained) disturbancesprincipally affect analogue-type circuits, whiletransient and impulsive disturbances interfereespecially with digital circuits.

* Stationary (but changing in magnitude) magnetic andelectric induction fields generally, are only significant inclose proximity to the sources and are easily countered byplacing sensitive equipment at a suitable distance fromthem. A notable exception is the case where some tens ofthousands of amps, i.e. short-circuit fault currents, flow in apower cable.

I

I

U equipment

I

equipment

2

I2

U

Appendix EMC - 3

EMCElectromagnetic disturbances couple readilywith cables in the common mode, particularlyat high frequencies (HF), since they act asradio antennae. Several kinds of couplingbetween neighbouring circuits can occur.The problems of common mode recurfrequently in EMC cases. A conductingenvironment is always good for EMC,due to it equipotential quality.Only disturbances in the differential mode canbe filtered locally, cable by cable.As indicated by its name, the common modeis common to all the cables of a givenequipment. Common mode problems at HFare particularly critical in an insulatedenvironment, or where the mass (the zero-voltage refernce for all electronic circuits) is"floating" with respect to earth (i.e. insulatedfrom the earth).A common mode voltage is alwaysdetrimental. If it cannot be reduced, it isimportant, at least, to prevent it fromdeveloping into a differential modedisturbance.

This is the role of insulated and/orsymmetrical connections.A galvanic insulation is only effective at lowfrequencies. A symmetrical connection, alsoreferred to as "balanced", can remaineffective up to high frequencies. Thedissymmetry of a differential connectionoriginates mainly from its end circuits. Animbalance at an end circuit can be caused byan electrical and/or geometric dissymmetry.In any case, a connection by simple coaxialcable to transmit signals at low frequencies isnot recommended.Correcting measures which may produceharmful secondary effects must be combinedwith other precautions in order that thesystem effectively counters the entire rangeof disturbances (LF and HF, of large andsmall amplitudes). The combination of thedifferent corrective measures (galvanicseparation, symmetrical connections andovervoltage protection) is referred to ascoordinated protection.

2.1.2 LF disturbances by conduction

LF disturbances include all types of parasiticinterference of which the range of significantfrequencies is lower than 9 kHz. Thefrenquency of 9 kHz is a conventional upperlimit, below which electrical phenomena maybe analysed in simple terms, using customaryequivalent linear circuit techniques, based onresistances, inductances (self- and mutual-)and capacitances. By definition, a LFdisturbance exists for a relatively "long" time(at least a hundred microseconds). Theenergy level of a conducted LF disturbancecan be considerable and is very easilymeasured.The impedance of a cable at very lowfrequencies is practically equivalent to itsresistance only. At several kiloherz mostcables of small cross-sectional-area (c.s.a.)and even at 50 Hz (for cables of large c.s.a.)the lineal inductance of a conductor is of theorder of 1 µH/m, and its impedance increaseslinearly with frequency. For example, large 1-core cables at 50 Hz, installed in trefoil, havea lineal impedance of approximately0.3 Ω per km. This feature is important whenconsidering harmonic frequencies in anetwork.Selecting a c.s.a. larger than 35 mm2 for aprotective conductor would effectively reducethe heating of the conductor when carryingfault current (since its resistance would belower) but would have a negligible effect onthe equipotential distribution: the inductanceof a cable being (as noted above) practicallyindependent of its c.s.a.

Interruptions (long or transitory)

An interruption is a total disappearance of thepower-supply system voltage. In the case ofa fault occurring on the HV network of apower-supply system, a consumer willnormally experience a "voltage dip"sometimes followed by a brief interruption.This interruption will occur only if the HVsystem is an overhead-line (O/H line) system,and the consumer is being supplied from thesection of line on which the fault hasoccurred. So-called "fleeting faults" onoverhead lines are very common, and consistof flashovers (of insulators) to earthed metalby momentary overvoltage due to lightning ora short-circuit through a large bird, or again,directly to earth through, for example, a wettree branch, etc.In more than 80% of these incidents such afault will disappear during the brief period ofautomatic interruption, and normal supply willbe restored.The automatic sequence for the elimination offleeting faults on O/H lines is included in theprotection scheme for the line. Theinterruptions in these schemes are normallylimited to less than 0.5 seconds. Anunderground cable supply network reducesthe number of interruptions to about 10% onlyof those of O/H line systems, butunderground cable faults are not self-clearing, so that lengthy shutdowns arenecessary to locate the fault and to effectrepairs.

4 - Appendix EMC

EMC

2. electromagnetic disturbances (continued)(continued)(continued)(continued)(continued)

Flicker

Flicker describes a condition of small butfrequently recurring voltage dips caused byloads which require relatively heavy currentfor brief and regularly repeated periods.The impedance of a LV network is made upmainly of cable impedance and theimpedance of the HV/LV transformersupplying the network. The greater the kVArating of the transformer, the lower itseffective impedance. In public power-supplysystems the flicker problem is more commonon rural systems, particularly at the end oflong lines. It is a problem on a line which issupplying arc furnaces, arc-welding machinesand, generally, where heavy loads arefrequently switched.Flicker creates an objectionable annoyancefor persons working under incandescentlighting. The effect is purely physiological, nodysfunctioning of electronic equipment willoccur due to flicker.Flicker is objectionable only where heavyoverloads and frequent switching arecombined, or where the impedance of thesystem is high. Standardized parameter limitsand a flicker meter are described in IECpublications 1000-3-3 and 1000-4-15.

2.1 disturbances by conduction (continued)

For industrial installations subject to flicker, amodification to the installation is sometimesnecessary. Among the possible correctivemeasures available, the most effectiveinclude: separate cables for heavy loadspreferably with each large load suppliedthrough an individual HV/LV transformer,division of the load, increase the time lags inautomatic control systems, reduction in thework-cycle rate, time-wise staggering andspreading of operations which requireimpulsive power demands, together with theinstallation of a static reactive-powercompensator. Technically, a reduction in thesource impedance is an excellent solution.In final general-distribution circuits at LV, the3-phase symmetrical short-circuit current isusually within the 500-5,000 A range. Inindustry, the short-circuit current at LV mayexceed 10 kA on a circuit of large c.s.a. closeto the source substation. This value never,however, exceeds 100 kA.

Fluctuations and voltage dips

A fluctuation of voltage is a rapid change ofsupply voltage not exceeding + 10% (thegenerally accepted limits at distribution level)during normal operation. A "dip" is a sudden

drop of voltage, caused principally byswitching loads which, at the instant ofenergization, require a greater current thanthe normal rated value, e.g. small-motorstarting currents, the switching on of largeresistive heating devices and incandescentlamps, etc. Such dips are transitory only, butare often more severe than those classed asflicker, generally exceeding 10%. The durationof a "dip" lasts from 10 ms to approximately 1 s.Voltage reductions which exceed 10% and1 s, due, for example, to starting large motors,or, as previously described, due to systemfaults, are simply referred to as "voltage drop"and the extent of the drop and its duration arespecified. Voltage fluctuations have littleeffect on electronic circuits generally.Sensitive, precision electronic-control devices,electronic calculators of early design, andelectronic (HF) fluorescent lighting tubes,however, may be adversely affected.A well-designed electronic device cantolerate, without difficulty, voltage fluctuationsup to + 8%.A voltage dip at a point on a HV system isgenerally due to a short-circuit faultelsewhere on the same network. The closerthe fault to the point in question, the moresevere the dip. The severity of the dip isdefined by two parameters: the magnitude ofthe drop as a percentage of the systemnominal voltage, and its duration inmilliseconds.Voltage dips are generally due to wind-blowndebris (tree branches, etc.), electric storms,or faults on the lines (broken insulators) oroccur on the installation of a neighbouringconsumer.

Faults on VHV (very high voltage)transmission lines are rare and are usuallydue to lightning, or to exceptionally severecold weather.The consequence of voltage dips (whenfollowed by an interruption) is a complete lossof supply to electronic (and power) devices.Relays will drop out and motors controlled byelectronic speed-variation and regenerative-braking devices will be deprived of brake control.Even if there is no supply interruption, a largeand long (up to 1 second) voltage dip maycause similar malfunctions.Means of countering these problems at theleast cost requires individual analysis of eachcase. To overcome the voltage-dip problem,many low-power electronic equipments haveindividual power packs with an autonomy ofseveral hundred milliseconds for 100% lossof supply voltage.For heavy-duty power supplies, the period ofautonomy amounts only to approximately20 milliseconds, the limiting factor being thesize of the energy-storing capacitors required.Rotating machines (motor/generators) have

most influenced beetween the three phase to phase voltages

time

10

depth(% de Un)

400 V

Ueff.

360 V

clearingtime

= 0,3 s

duration= 0,4 s

fig. EMC-4: caracteristics of a voltage dip.

fig. EMC-3: number of variations per

minute.

3

2

1

0,5

0,3

.5 .7 1 10 100 1000

∆U/Un in %

Appendix EMC - 5

EMC

Unbalance

The amplitude of an ac voltage is expressedby its rms (root-mean-square) value. Thevoltage between a phase conductor and theneutral is referred to as the phase voltage,while that measured between any two phasesis called the line voltage. The line voltageequals etimes the phase voltage, on anormally balanced 3-phase system(e= 1.732). A 3-phase system may bedefined simply by the amplitude of 3 voltages,either line or phase values.In order to define a sinusoidal system whichis in an unbalanced state, however, thevalues of current and voltage of each phaseare then, in the general case, the sum of3 rotating vector components. The threecomponents of each phase are known as:c the positive phase-sequence componentc the negative phase-sequence componentc the zero phase-sequence component.A balanced 3-phase system is composed ofpositive phase-sequence components only.An unsymmetrical system is said to beunbalanced; its negative- and zero-phasesequence components are generally bothpresent, together with the positive phase-sequence component.A common cause of imbalance is that ofdifferent levels of loading on the 3 phases.Unbalanced loading results in unbalancedvoltages being applied to 3-phase motors.Increased losses occur in the rotors of themotors, and in cases of excessive imbalance,motors can be destroyed by overheating.Single-phase (line-to-line) loads are notnormally adversely affected by imbalance.Small degrees of imbalance (0.5-1%) areinevitable on LV 3-phase 3-wire networks,and up to 2 or 3% can be tolerated for severalminutes by all loads.When an imbalance of voltage is consideredto be excessive (> 2% for example), it isadvisable to correct the balance of phaseloading. Where it is not possible to improvethe balance, the situation may be eased byincreasing the fault level at the circuitconcerned by changing the supplytransformer.An average HV/LV distribution transformer(> 100 kVA) has a short-circuit voltage of5-6%. Special transformers are available withinterlaced windings which limit the leakagereactance to give a short-circuit voltage ofapproximately 2%.A low short-circuit voltage effectively means alow source impedance (with higher fault-current level) a situation which improves thevoltage balance, and (incidentally) improvesthe form of the voltage wave (if it happens tobe distorted) by reducing the harmoniccontent of the wave. A modern method ofimproving a condition of imbalance, thoughpresently rather costly, is to install a staticcompensator. It consists of a system whichstores energy in an inductor or capacitor, andrestores this energy to the system at theappropriate instants.An active filter constitutes one of thepreferred solutions for limiting disturbancesgenerated by arc furnaces during the start-upphase.

Frequency variations

The European network performs, in practice,as an infinite system as far as frequencystability is concerned, in that load changes donot sensibly affect the frequency. On smallerprivate systems, and especially on singlegenerators, where the rotational inertia issmall and the regulating system of the primemover is generally rudimentary, the frequencywill vary (within reasonable limits) each timethe load changes abruptly. Diesel engineprime movers are less stable, in terms offrequency, than turbines. Frequencyvariations do not unduly disturb electronicequipments. Converters based on currentchopping principles are insensitive tofrequency changes. All modern devices andcomponents should be capable of correctperformance during frequency changes of+ 4% throughout a 10 minute period.Only very large systems with transformersoperating at the limit of saturation may, whenthe system voltage is at its maximum, besubjected to overheating by a long-term low-frequency condition. AC motors (locked to thefrequency) will experience speed variationscorresponding to those of the frequency. Onthe other hand, the inertia of motors tends tosmooth out other sudden disturbancesoccurring on a network.

Harmonics

Any non-linear load (fluorescent lamp, Graetzbridge, arc furnace, etc.) takes a non-sinusoidal current from the network. Such acurrent is composed of a sinusoidalcomponent at the frequency of the systemand is known as the fundamental component,together with other sinusoidal componentswhich are whole-number multiples of thefundamental frequency. These latter arereferred to as harmonic components.Conventionally, harmonics up to the rank of40 only are considered in power systems,i.e. 2 kHz for 50 Hz systems and 2.4 kHz for60 Hz systems. Supplies to electronic circuits,power regulators based on Graetz bridge,and fluorescent lighting equipment, are rich inharmonics.Distortion of a voltage waveform is onerousfor associated equipments; it is expressed asa percentage. It is proportional to theharmonic content of the current and to theimpedance of the source. The effect ofdistortion is to increase the heating losses inmotors. In an ADP environment, a distortionof 5% may be considered to be normal. Allelectronic components can tolerate a globalfactor of distortion including possible inter-harmonics of at least 8%. An inter-harmoniccurrent has a frequency which is not a whole-number multiple of the fundamental (i.e.system) frequency. A distinction is madebetween "true" inter-harmonics generated atdiscrete frequencies, and those forming partof a continuous spectrum.Even-numbered harmonics are generatedonly by asymmetrical rectifiers and loadcurrents which contain a dc component. A dccomponent can readily saturate a power-supply transformer. Most non-linear loads(satured transformers, fluorescent tubes,power supply circuits which use current-chopping techniques, etc.) only generateodd-numbered harmonics.

sufficient autonomy to cope with voltage dips.Finally, uninterruptible power supply units cansuppress voltage dips, and maintain thepower supply during a period of completeinterruption.

6 - Appendix EMC

EMC


2.1 disturbances by conduction (continued)

Balanced 3-phase loads supplied froma 3-phase 3-wire system (i.e no neutral wire)do not generate 3rd harmonic currents ormultiples of 3rd harmonic currents. A Graetzbridge or a hexaphase thyristor-controlledregulator behave as current generators,which are practically independent of thevoltage distortion.Third harmonic currents and multiples ofthem (known as "triplen" currents) whengenerated in the phases of a 3-phase 4-wiresystem, present a particular problem in that,being of zero-phase sequence (i.e. in phasewith each other), they add arithmetically andcomplete their circuit through the neutralconductor. The current in the neutralconductor due to this cause is at 150 Hz andcan exceed, in unfavourable circumstances,the current flowing in the phase wires. Thisaspect should be borne in mind whenconsidering supplies to ADP loads and loadsmade up of fluorescent lighting tubes. If theneutral is not distributed, i.e. the system is3-phase 3-wires, then 3rd harmonic currentscannot flow. A transformer with a delta-connected primary winding, however, allowsthe circulation of 3rd harmonic currents, afeature that practically negates the distortionof the LV voltage wave, which wouldotherwise contain a large 3rd harmoniccomponent.Reduction of the source impedance is notalways effective in reducing distortion :power-factor correcting capacitors can causea problem, if, together with the sourceimpedance (which is predominantly inductive)the combination should form a resonant (orpartially resonant) circuit at one of theharmonic or inter-harmonic frequencies.The parallel combination of inductivereactance L and capacitive reactance C canpresent a high impedance at its resonantfrequency, particularly at low-load periods.The distortion can therefore becomeexcessive due to the amplification effect ofthe resonance.A harmonic filter is a series LC circuitconnected in parallel with the source and actsas a short-circuit to currents at its resonantfrequency. The capacitor and inductor mustbe capable of carrying the maximum value ofthe harmonic current, while the capacitormust withstand the elevated harmonicvoltage plus the system normal-frequencyvoltage. The inductor is adjusted to cause thecombination to resonate at the exactfrequency of the harmonic in question. It mustnot saturate or overheat. The rating of aharmonic filter varies according to the size ofthe installation from some kvar to severalMvar.A problem with harmonic filters is that theirresonant frequencies change from one pointon the network to another due to cableinductance. The longer the cabling, the lowerthe frequency at which the filter resonates.It is advisable moreover, to check that thecurrents in a number of paralleled filters aresatisfactorily shared.A modern means of limiting distortion is bythe use of active filters. These filters areinverters based on PWM (Pulse WidthModulation) techniques, together withreactive-energy storage.

The role of an active filter, triggered by anerror function, is to inject into the network aharmonic current exactly equal, but in phaseopposition, to that produced downstream.The principle of operation is comparable tothat of a static compensator, but at aconsiderably higher chopping rate. An activefilter can be called upon to compensate notonly first (i.e. lower) harmonics, but alsoreactive power (var) and flicker ; it is simply aquestion of dimensioning.

Overvoltages

Overvoltages that affect industrial-supplypower networks in the differential mode(between phases) can occur for numerousreasons.The energizing of a bank of capacitors cangenerate an overvoltage transient associatedwith an energy level of several hundredJoules. The inductance of the system and thecapacitance of the bank behave as a seriesLC circuit which oscillates transiently at itsnatural frequency, typically below 1 kHz.The value of the first peak (the sum of thetransient and system peaks) can reachalmost twice the value of the power systempeak voltage, i.e. the transient can have apeak value almost equal to the system peak.A second cause of overvoltage may occur onthe blowing of a wire-type fuse. A suddenrelease of the magnetic energy stored in thesystem inductances can be as much as athousand Joules, which, when converted intoelectrostatic form in the system capacitances,can raise the voltage sufficiently to damageassociated equipment. The effect isconsiderably less with cartridge fuses or witha circuit breaker.Any manœuvre on a power transmissionsystem (opening or closing a circuit breaker,isolating switch, etc.) results in an operationaldisturbance. The energization of atransmission line is characterized by a wavefront of voltage which (at approximately thespeed of light) propagates and reflects alongthe line to produce a voltage-doublingphenomenon.The frequency of this highly-damped short-duration oscillation varies between 10 kHzand 1 MHz. The phenomenon is related tothat of energizing a capacitor, but thefrequency in this case is much greater(depending on the length of the line) while theenergy level is lower. The risk of damage toequipment from this phenomenon is muchless than that from overvoltages of longerduration, but the risks of malfunctions aregreater.Overvoltages have little effect onelectrotechnical equipments, but can disturb,weaken and even destroy electronicequipments. A LF overvoltage in commonmode has no effect on supply circuits that aregalvanically isolated, provided that theisolating dielectric can withstand the voltagestress without rupturing. At HF, principally byits common-mode component, anovervoltage can disturb (i.e. interfere with)sensitive electronic systems. The solution issimple: filter each equipment with respect toits own mass (metallic structure of theequipment).

Appendix EMC - 7

EMCThe installation engineer has practically onlyone way to protect the installation againstovervoltages, and that is to install overvoltagelimiting devices on the supply-circuitconductors. Overvoltages occurring on publicLV distribution networks are lower in energythan those occurring in heavy-currentindustrial networks: the energy on publicnetworks rarely exceeding 100 Joules.The only really dangerous case is that of alightning stroke on a line close to theinstallation.

New overvoltage surge arresters based onthe use of varistors of high energy-dissipationratings allow an effective protection of all LVsystems and equipments downstream of thepoint of arrester installation.Failure of the zinc-oxide varistor will cause athermal fusing element (which is connected inseries with it) to blow, and open the circuit,thereby avoiding a short-circuit to earth via afaulty or damaged arrester. The cable fromthis device must be connected by theshortest possible route to the mass of thedistribution board, i.e. the common earthingbar, and not to the earthing electrode, whichis generally too far away (see Sub-clauseL 1.4).

2.1.3. HF disturbances by induction

At HF i.e. conventionally above 1 MHz,interference phenomena becomeconsiderably more complicated. Powerconductors become efficient antennae,electromagnetic fields, even when weak,produce considerable interference, all cablesare affected, and some may resonate, etc.HF phenomena are severe, frequent, difficultto analyse and are cause to reconsider theestablished practices in electronics cableinstallation.The inductance of cables becomes more of aproblem at HF than at low frequencies. Thelineal inductance of any conducting structurefollowing a sensibly straight route isapproximately 1 µH/m. Furthermore, aninterconnection of a length exceeding athirtieth (1/30) of a wavelength becomespractically incapable of ensuringequipotentiality between the twointerconnected masses. Beyond λ/30, aconductor becomes an effective radiatingantenna but, if radiating, it fails to performcorrectly as an equipotential conductor.The wavelength λ corresponding to afrequency of 1 MHz is 300 metres. Thedistance between any equipment and themain earth bar being generally greater than10 metres, one may deduce that the natureand the quality of the earthing is of noconsequence at frequencies exceeding1 MHz. A simple dictum: a large conductor isgood, but a short conductor is better.HF disturbances by conduction in commonmode through cables are, in fact, consideredto be the principal problem for EMCspecialists. The reduction of common modedisturbances at HF through cables can beachieved by one or more of the followingthree stratagems:1 - attenuation effects: close interconnection(networking) by equipotential conductors of"masses" and/or screened cables2 - filters between conductors andmechanical mass of each equipment3 - ferrites on "problem" cables.An electronic circuit, e.g. a card supportingchips, etc., should never be allowed to "float"relative to its conducting envelope, acondition which is to be avoided at all costs inthe presence of HF interference. The natural(so-called "stray") capacitances of the cardcomponents, less than a pico-farad, can besufficient to cause interference with anelectronic circuit. To limit rapidly varyingvoltages between an electronic circuit and itsenvironment, the connecting of the filter 0 V(reference voltage) terminal to an envelopingmetallic housing, connected to earth or not, isan excellent preventive measure.

HF spikes

The range of frequencies which presents thegreatest difficulties, both in radiation and inprotection against the radiated energy, is theVHF band from 30 to 300 MHz, also referredto as the "metric" band. Almost all electricarcs, sparks, electrostatic discharges andstarting contacts (such as dry contacts,starting contact for striking an arc inelectroluminescent tubes, operation of circuitbreakers and other switching devices on HVsystems) generate impulses (spikes) whichare conducted in common mode andradiated. The radiation spectrum covers therange of the above-mentioned VHF band.The amplitude of the HF current spikes canattain a peak of several tens of amperes.Digital circuits are particularly vulnerable tosuch spikes. Respecting the standard forimmunity IEC 1000-4-4 is a highly-recommended means of achievingsatisfactory protection and EMC of aninstallation.Maintained HF disturbances

Frequency converters, electronic speedcontrollers, Graetz bridges and electric-motorcommutating brushes also generate commonmode HF disturbances. The peak value ofthese disturbances can reach and evenexceed 1 ampere. One solution is to install anefficient filter at the supply-source and/or atthe disturbed equipment. Another solution isthe use of power cables that include ascreen, which is earthed at both ends.For sources of heavy interference, it isrecommended to form a network ofequipotential interconnections of all massesin the neighbourhood of the offending source,in particular all metallic cable ways, ducts,trays, etc.

8 - Appendix EMC

EMC

2.2.1 LF magnetic fields

At low frequencies, only the magnetic fieldmay cause problems. Whether it is impulsive(short-circuit, lightning, electronic flash...) ormaintained, the field H is generally producedin close proximity to the affected equipment.Measurement of the field strength requires anoscilloscope and a loop probe only. Themagnetic field at low frequency does notpropagate, but remains in the close proximityof its origin (a transformer or an inductionmotor, for example) and its field strengthinitially decreases rapidly with distance fromthe source according to 1/D3.At greater distances, the rate of decrease isslower and approaches 1/D2. This latter valueis often used when considering the fieldsurrounding busbars or an overhead line. Themagnetic-field strength of a rectilinear currentwith a return path at infinity (such as that dueto lightning) decreases according to 1/D.Severe sources of magnetic fields are thezero-phase sequence currents in supplycables of a TN-C scheme. Loops formedbetween the phase conductors and currentsdiverted from the neutral conductor (throughequipotential bonds) are sometimes very


2.2 radiation

Electrical energy propagation is not onlyconfined to conductors. It can also propagatein space without material support. Suchpropagation is referred to as electromagneticfields or waves, or Hertzian waves. Such awave is made up of an electric component Ein volts per metre, and a magneticcomponent H in amperes per metre.These radiated fields, when encountering a

conductor (which acts as a receivingantenna), give rise to minute emfs andcurrents in the conducting material, i.e. in theform of disturbance by conduction. For cablecircuits, these disturbances are in commonmode. It is therefore possible to protectagainst these radiated fields by means of aFaraday cage arrangement or by (very often)low-pass filters.

large; such currents can amount to severalamperes. The undesirability of the TN-Cscheme (in buildings) for this reason is shownin fig. F14 Chapter F, of the main text. Duringa short-circuit fault, the disturbance isevidently greater to a degree that depends onthe fault-current magnitude.The most common consequence of a LFmagnetic field is a distortion of the image of acathode ray tube (jumps and wave-likemovements of the image, and even changesof colour). A magnetically unscreened CRT(cathode-ray tube), an electron miscroscope,a mass spectrometer or a magnetic readinghead, barely tolerates 1A/m at LF. Moreover,the "stray" loops formed through equipotentialconnections to masses are associated(naturally) with corresponding voltages.Magnetic screening from/of a magnetic fieldis very difficult at frequencies less than10 kHz. The easiest solution is simply toplace the sensitive equipment out of range ofthe offending field. Screening the sensitiveequipment with a thick magnetic shield canreduce the field strength by a factor of theorder 10.

2.2.2 HF electromagnetic fields

At high frequencies, the E and H fields uniteto form indivisible electromagnetic waves inspace. At more than a sixth of a wavelengthfrom a point source, the ratio E/H tends to avalue of 120 π = 377 ohms. It is sufficient,therefore, to give the value of one componentin order to deduce the field strength.Numerous industrial, scientific or medicalapparatuses use radio frequencies, mostoften in the 1 MHz to 3 GHz range. Radiotransmitters have power-radiation capabilitiesranging from several milli-watts for radio-control devices, to several mega-watts peakfor radar systems.Walky-talkies, which can be used fortransmission very close to electronicequipment, are sources of disturbance,particularly for low-power analog circuits.An effective way to reduce radio-transmitterfield strength, as "seen" by sensitiveelectronic equipments, is to use antennae asremote from the equipments as possible, andlocated at the greatest height attainable.Since this principle cannot be applied toportable transmitters, their use must berestricted to areas sufficiently remote fromsensitive equipments to ensure trouble-freeoperation of the latter.

Electronic equipments are rarely affected bya field-strength of less than 1 volt per metre.Field strengths exceeding 10 volts per metre,however, very often cannot be tolerated. Therange of frequencies giving the most severeeffects is, again, in the VHF band.At HF a common-mode current in a cablealways produces a radiated wave. Thereciprocal case is also true, i.e. the arrival of aHF wave will produce a common-modecurrent in a cable. The methods of protectionagainst HF fields are the same as thoseadopted against disturbances by conductionat the same frequencies.The antenna effect of cables carrying HFcurrent by coupling in common modeconstitutes the principal problem in EMC.

Appendix EMC - 9

EMC

3. cabling of equipment and systems

In order to cable an electronic systemcorrectly or to correct an unsatisfactoryinstallation, it is often sufficient to apply somesimple elementary rules. From experience,the most important factor is a clearunderstanding of the phenomena and therecognition of their limits. The strictobservance of traditional rules for correctinstallation and cabling has becomenecessary. This is the price to pay for the

achievement of EMC in modern electronicsystems. Many practices which aresatisfactory at LF have proved to be poor oreven catastrophic at HF. Certain cablingoptions can be chosen with confidence.The interconnection of all the non-functionalearths of a single site is one example.Factors which are always favourable shouldbecome standard practice.

3.1 earthingThe expressions "earth", "earth electrode","earth plate", "earth rod" all refer to aconductor which is buried, and in intimatecontact with the soil. The word «mass» refersto metal parts of equipment (electrical or not -for example, water pipes) which, undernormal conditions, are not intended to carrycurrent. Bonding conductors, which are usedto interconnect masses are also referred toas "mass". Although all masses in normal LFinstallation practice are connected to earth,the two words, "earth" and the above-notedequivalents should not be confused with"mass". "Mass" is commonly called "ground"in some countries.

3.1.1 the role of earthing

The basic role of an earth electrode is tomaintain all masses in an installation at avoltage close to zero, whether the powersource is earthed or not. This is achieved in aproperly designed installation, regardless ofwhether a faulty condition (which wouldotherwise raise the voltage of the installationmasses) occurs in the installation circuits oron the power-supply network, or othersources external to the installation.The role of earthing, therefore, is that of theprotection of persons against the dangers ofelectrocution. The severity of an electricshock is a function of the current whichpasses through the body, and equallyimportant is the path of the current flowthrough the body. Established IEC rules ofprotection against electric shock set safelimits of voltage (referred to as conventionalvoltage limits) above which masses areconsidered to be unacceptably dangerous.For normal 50 Hz or 60 Hz power systems,these values are 50 Vrms for dry locationsand 25 Vrms for wet locations, for examplebathrooms and laundries (see Chapter L ofthe main text for more details).It is recognized that a low contact resistanceof an earth electrode with the mass of theearth cannot always be obtained.Furthermore, its value is rarely constant,depending largely on soil humidity (and soprone to seasonal changes). An essentialfactor in maintaining safety to personnel inthe event of a high earthing resistance, is thatof the equipotential concept. If, for example,all masses are at a common (even normallydangerous) potential, and the earth below thebuilding is at a similar potential, a person cantouch any or several masses at the sametime without danger. This is why electricalappliances with long leads (hedge trimmers,lawn mowers, etc.), which allow the user toleave the equipotential environment of thehouse, must be of Class II insulation level(i.e. doubly insulated).So-called normal leakage currents (noinsulation is perfect) also include the minutecapacitive currents of the wiring to earth.

These currents, and short-circuit-to-earth faultcurrents, flow mainly through the protectiveearthing PE conductors (coloured with yellowand green stripes) and finally back to thesource substation, via the earth (TT system)or via the earth path and (mainly) through theneutral conductor in parallel (TN system).Since, in the TN case, practically all of thefault (and leakage) currents return to thesource via the neutral conductor, theresistance of the installation earth electrode isnot of primary importance (unless lightningarresters are to be connected to it).For the protection of electronic equipments, itis strongly recommended that common-modecurrents entering the building from externalcables be diverted to earth at the point of entry.A simple galvanic isolation is ofteninsufficient: the overvoltage withstandcapability of a galvanic isolation transformeris typically less that 10 kV. This value isinsufficient on days of intense electricalstorms.The installation of non-linear voltage-limitingdevices then becomes a necessity.It is important that all incoming metallic pipes,ducts, trunking, etc., be connected to earth attheir entry into the building. This policy canavoid the circulation of currents (from outsidethe building) in the conductors bonding themasses.The installation of overvoltage-protectiondevices must be carried out with theminimum possible common impedancebetween the external circuit and the circuit tobe protected. The length of conductor inseries with the voltage limiter must,consequently, be the shortest possible.The residual voltage "seen" by the protectedequipment is then independent of theimpedance of the earth.Even with a "bad" earth, it is possible toprotect an equipment effectively againstexternal overvoltages: it is necessary andsufficient to connect the voltage limiter to themass of the equipment using the shortestpractical length of cable.

fig. EMC-5: a voltage limiter must be

connected to the mass and not to earth.

protectedelectronicequipment

electric-power line

earthing conductor

voltage limiter

yes

no


10 - Appendix EMC

EMC

3. cabling of equipment and systems (continued)(continued)(continued)(continued)(continued)

3.1 earthing (continued)When a voltage limiter is correctly connectedto the mass, the impedance of the earthelectrode is immaterial.A direct lightning stroke on the supply lineclose to the installation requires thedissipation of (typically) 10 kA to 100 kA ofstroke current, most of which passes to earththrough lightning arresters on the lineexternal to the installation. Overvoltageswithin an installation where external arrestersare provided rarely exceed 6 kV due toatmospheric causes.

fig. EMC-6: the entire equipotential

"cage" will be at a high absolute potential

during the brief flow of stroke current.

For transmission and distribution lines at highvoltage, the fault current for one phase toearth returns to the source through the earthand (where provided) through the shieldingconductors above the phase conductors ofthe lines.The provision of an equipotential condition onthe surface of the ground at the base oftransmission towers and, more importantly, atsubstations (the source of the fault current) isa primary concern of design engineers. Theprinciple of equipotential bonding is identicalto that required for a low-voltage installationin a building.A functional earth means an earth electrodewhich is designed to pass load currentthrough the earth, i.e. the earth path acts asone of the circuit conductors. There areseveral installations around the world thatutilise this method. In some countries, d.c. isused in this (economic) way to operate aservice of facsimile transmission.It may be of interest to note that whentelephone cables (which use paper insulationon the conductors) have a high degree ofleakage current and consequently doubtfulsymmetry, a low earth resistance allows thequality of the transmitted signals to bepreserved. Although the magnitude oftelephone signals is low (millivolts to lessthan 1 volt), the quality of modern cablesovercomes the constraints of a "good" earth.To summarise, the protection of persons doesnot depend directly on a low value of earthresistance*; it is rather the establishment of acondition of equipotential between massesthat is of prime importance. Thus, an aircraftin an electric storm presents no danger to thepassengers, who are in a metallic envelopewhich is (within a few volts) perfectlyequipotential. For persons or animals, thedanger is not the magnitude of the absolutepotential, it is the difference of potentialbetween metal parts which can be touchedsimultaneously that is dangerous.

An electronic equipment is not affected by thevalue of earth resistance. At worst, there is arisk of exposure to overvoltages from anexternal cable, if its protection is insufficient,or is badly cabled. So that the role of themasses is essential, and more important thanthat of the earthing. The only requirement fora satisfactory performance of the electronicequipments is a high degree ofequipotentiality.It is evident that two earths are always lessequipotential than one. Any separate earth,even if said to be «interference free» isalways detrimental to equipotentiality, and soto the safety of persons and to thesatisfactory functioning of interconnectedequipments. Two non-functional earth-electrode systems on a common site shouldalways be interconnected.In practice, care must be taken that there isno touch-voltage existing when working on anelectronic equipment interconnectionbetween two buildings (video, access control,local network, information technology (ADP)devices...) if the earthing systems of the twobuildings are not solidly interconnected. It is,as already noted, not possible to be sure ofthe equipotentiality of two separate earths.* This statement is not true in certain circumstances, notablyin rural areas, where a small transformer supplies anisolated community, for example. The neutral earthelectrode must necessarily have the lowest possibleresistance in such cases. If not, potential gradients on thesurface of the ground can be dangerously high in the vicinityof the electrode during an earth fault.Animals are frequently killed due to this cause.

direct-stroke current

lightning protection conductors

internal equipments must be maintained at

equipotential

Appendix EMC - 11

EMC

3.2 masses

The majority of malfunctions inelectromagnetic devices, sometimes wronglyimputed to software problems or human error,are found to be due to an insufficient level ofequipotentiality between interconnected units(probes, cards, actuators).There are two differences between a buriedconductor and a mass conductor. A buriedconductor will dissipate its common modecurrents, but it is always too far from theequipments to be effective at HF.A mass conductor above ground levelpresents two essential virtues to the goodperformance of electronic systems, viz: it isphysically close to the circuits, and it isaccessible.The equipotentiality of equipments and theirmasses is a functional objective.As long as interference signals circulate inthe masses and not in the electronic circuits,they are harmless. On the other hand, if themasses are not all equipotential and areconnected to earth in star*, for example, HFinterference currents will circulate throughany available path, i.e. via signal cables.Some circuits will therefore be subjected tointerference and even be destroyed.Networking the mass conductors to form aclosely-connected low-impedance bondingsystem is the only economical way to ensurea satisfactory level of equipotentiality to installall sensitive equipments in a "Faraday cage"arrangement (a room enclosed in a mesh ofconductors) would be technically ideal, but isgenerally not economically justified.By definition, a "mass" is any conductingmaterial which can be touched by any part ofa human body, that is not normally alive, butwhich may become alive as the result of afault. Two masses which are accessible andwithin human reach must present a potentialdifference, under any conceivable faultcondition, that does not exceed the IEC-recommended conventional voltage (UL)

safety limit (of 50 Vrms in dry locations and25 Vrms in wet conditions for ac systems).These values are the maximum allowed thatcan exist indefinitely in specified conditions ofexternal influences.A dangerous touch voltage can arise during afault if the resistance of the equipotentialconductors is not sufficiently low. In somecases, it is necessary to install supplementaryequipotential conductors in parallel withexisting conductors to satisfy the UL criterion.It should be noted that access to two (ormore) masses is illegal, even if they areassociated with different installations, if theyare connected to different earthing systemswhich are not interconnected.The respect of safety rules is obligatory butnot sufficient, in themselves, to ensuresatisfactory EMC of an installation. In fact, therisk of electrocution only exists by a voltageof a high value and relatively long durationappearing between adjacent masses.An electronic equipment is sensitive to anextensive range of frequencies, or to verybrief impulses. An electrostatic discharge, forexample, is generally of no consequence forits source, but it could be catastrophic for anelectronic device. Normal earths, commonlyconnected in «star» (i.e. radial) configurationfor example, guarantee the safety of personswhen the relevant standards are respected,but not the satisfactory operation of aninstallation which includes sensitive electroniccomponents.It is certain that more and more electronicequipments are, or will be, connected to otherapparatuses and devices for the exchange ofinformation. The best way of ensuring asuccessful and durable installationperformance is to establish a high degree ofequipotentiality throughout the entireinstallation.* i.e. by one of a number of conductors radially connected tothe main earth bar, the ensemble resembling a "star".

3.2.1. loops of mass and between masses

A loop of mass is the area included betweena working cable (metering cable, controlcable, power-supply cable, local-networksystem cable) and the mass conductor(generally the nearest PE conductor). Thereare, therefore, as many loops of mass asthere are cables. This is inevitable, whetherthe conductors are galvanically isolated ornot. A galvanic isolation reduces thecirculation of LF currents without, however,reducing the area of the loop. A loop canoscillate strongly at HF, so that large-arealoops constitute the major problem in EMC.

fig. EMC-7: there is an inevitable mass

loop per cable.

If a current circulates around a mass loop,such a current in common mode may eithersuperimpose «noise» (interference) on usefulsignals (in differential mode, by conversionfrom common mode to differential mode) ordisturb the electronic circuits at eachextremity.The risk is equally present for radiation from aloop as for reception of interference by theloop. The output stages of electronic circuitsare as sensitive to interference as the inputstages, and are more difficult to filter.The areas enclosed by mass conductorsmust not be confused with those referred toabove as "mass loops". It is preferable toallow parasitic currents to propagate in themasses rather than in the signal cables.These loops between mass conductors arecalled "loops between masses".

fig. EMC-8.

equipment 1 equipment 2

signalcable

massloop

nearest mass conductor


signalcable

mass conductor

loop between masse

mass conductor

12 - Appendix EMC

EMC


3.2 masses (continued)If two neighbouring masses are notconnected together, the differential potentialdifference between them may be significant.A direct connection from one to the other willalways improve the equipotential condition.At least, the masses of all equipments whichexchange data between them should beinterconnected by mass conductors. An evenmore certain way of improving the state ofequipotentiality is to interconnect allequipment masses, whether they exchangedata or not."Mass loops", also called "ground loops",should never be confused with «loopsbetween masses». A mass loop is neverfavourable, and its area must be reduced tothe minimum attainable, to reduce as far aspossible the interference effects of disturbingfields. On the other hand, it is always goodpractice to increase the number and reducethe areas of loops between masses. fig. EMC-9.

3.2.2 unity of the mass networkThe mass must be unique to be equipotential.There are three methods of connectingmasses which preserve this unity.

1 - Earthing connections in "star": eachequipment has its own earthing cable, whichterminates with all other individual earthingcables on a unique earthing bar.

fig. EMC-10.

The justification of such a philosophy is over-simple: when an equipment develops a leakagecurrent to mass, the remaining equipmentsare assumed to remain at earth potential.But "earth" potential has no real meaning inpractical electronics, all potentials are relativeone to another, the concept of absolute zeropotential (i.e. "remote earth") is abstract.It is often assumed that the "star" configurationof earthing overcomes the problem ofcommon impedance. It is, in fact, quite theopposite! Earthing in the "star" configurationincreases the common impedance (that is,forms a point of common coupling) betweeninterconnected equipments.

fig. EMC-11.

Earthing in "star" can create a commonimpedance between two interconnectedequipments.It is sometimes also assumed that the "star"system of earth connections suppresses themass loops. Between two interconnectedequipments, it is evidently not the case; thearea enclosed by the mass loop can, in fact,be considerable. An electromagnetic fielddue, for example to a lightning discharge, willinduce voltage in the mass loop greater thanthat occurring in any other method ofearthing.

fig. EMC-12.

This long-established "star" earthing methodis now only possible for an equipment whichis, and will remain, isolated from any other.The method can be suitable only forelectronic analog systems (as opposed todigital systems) with floating sensors, and theelectronic circuits completely isolated fromany other. Such cases are becomingincreasingly rare.With the generalisation of information-datatransmissions over great distances, localnetworks, shared peripherals and, in general,the exchange of signals betweenequipments, the "star" method of earthingmust be abandoned. Moreover, even if theearth connection of each equipment by anindividual conductor is not detrimental, it stillremains a costly method that requires largeamounts of copper and many hours ofinstallation work.The only reasonable application for the "star"arrangement of earthing (in fact, connectionto mass) is the connecting cable between anequipment and power-supply socket-outlet, orthe nearest distribution board. Thus, in anADP environment, it is reasonable to use thegreen-and-yellow PE conductor of the power-supply circuits to connect each equipment to

equipment 1 equipment 2greater immunity

againstradiation fields by

reductionof area of the mass loop

greater immunity from conducted interference by multiplication and reduction

in area, of loops between masses.

earth cablesinevitably long

authorized method, but costly and not good for EMC, particularly for interconnected equipments

earth cables radiating from the main earth bar, figuratively similar to a star

disturbanceon cable

high value of differential potential

difference (d.p.d.)

high impedance if the conductor is long


I

I

PE PE

Z

signalcable

mc


signalcable

large area

strongd.p.d.

electro-magneticwave

The more this policy is developed, the moreeffective is the resulting state ofequipotentiality, both at LF and at HF.Connecting the masses to a mesh ofinterconnecting mass cables is alwaysbeneficial, regardless of the nature of theequipments concerned.

Appendix EMC - 13

EMCthe general distribution board, located in theroom. From the common earth bar of theboard, a single protection cable is taken tothe main earth electrode for the installation.This conductor can be common to otherdevices, and may, with advantage, beconnected to neighbouring masses.

fig. EMC-13: a good earthing-system

arrangement.

Even if a strong source of interference isinstalled in the same environment assensitive equipment, a separate earthingsystem for the sensitive equipment isdetrimental and not recommended. At themost, it is desirable to supply the twoincompatible systems on separate cablesfrom the power-supply network. In any case,mesh-connecting the masses is favourable.Such a mesh of PE conductors has the meritof avoiding involuntary loops which canbecome catastrophic if not successfullycountered."Star"-system earthing can be accepted onlyfor low-frequency installations that are, andwill remain, independent and isolated fromany other installation.2 - Connection to the nearest PE

conductor: a unique protection (PE)conductor, associated with severalequipments.

fig. EMC-14.

By using this cabling scheme, the mass loopshave a small area and the commonimpedance between interconnectedequipments is less than that with a "star"-connected earthing scheme. This economicmethod is also recommended for safetyreasons. It is easy to prove that the touchvoltage between two masses connected tothe same PE conductor remains less than theconventional (UL) value. The risk of using thesame PE conductor for earthing two systems,one "noisy" and the other sensitive, is notnegligible. Although the low impedance of thePE conductors and the good level ofimmunity to LF interference in common modetogether limit the risks, the HF currentsgenerated by strongly polluting sources(power converters in particular) cannot bedissipated efficiently by a single PEconductor. In such cases, it is necessary toinstall additional (supplementary) PEconductors in the form of a meshed network.

3 - The shortest connection to the nearest

mass.

This third method of connection to thenearest mass is better than those previouslydescribed. It is based on the meshconnection of masses. The areas of massloops are reduced to the strict minimum andthe degree of equipotentiality of the massesis excellent.

fig. EMC-15.

Note: concerning the safety of persons, thistype of local connection is not generally asubstitute for PE conductors. It is appropriate,therefore, to superimpose the methods 2 (oreven 1) for the safety of persons, and n° 3 forEMC.Mesh connecting the masses is even moreimportant as the area covered by theinstallation increases, with longinterconnecting cables, or when theequipments are divided between severalstoreys.The mesh connecting of masses does notreduce the favourable policy of supplyingsensitive equipments by different supplycables than those used to supply stronglypolluting loads. However, the use of differentsupply cables does not signify "star" -connected masses.The lengths of protective conductors (PE)mean that, at HF, their impedance isgenerally too high to effectively improve theequipotential situation. For example, a PEconductor 100 metres long is incapable ofpassing significant currents at frequenciesexceeding 100 kHz.The PE conductors and earthing conductorsalone are not sufficient to ensure the EMC ofan installation. Additional conductors andshort inter-mass connections are alsonecessary.PE cables, even long, and short-circuited atone end by a mass grid, function correctly atLF for the protection of persons. Nointerconnection between masses and noprotective conductor should ever beremoved, even if they appear to havebecome redundant following a carefulmeshing (close interconnection ofequipments and other masses to form a"mesh") of all adjacent masses.A PE conductor is not to be considered as anearthing conductor but as a "bonding"conductor, or "earth-bonding" conductor, theprincipal function of which is to ensure that UL(i.e. the maximum allowable touch voltage) isnever exceeded.Furthermore, there should never be morethan one earthing system per installation (i.e.per site) similarly for the mass system whichmust be unique, and connected to the uniqueearthing system. If this policy is not adopted,then EMC problems will certainly beexperienced via the inevitable links betweenadjacent installations (entry controls, video,alarms, safety measures, etc.).

PE conductor forother installations


earth bar

sensitiveequipments

local mass

main earthterminal

earth-electrodesystem

a single PE conductor

structures of adjacent masses(mass grid, conducting false floor, cable trays,

ducts, troughs, etc)

14 - Appendix EMC

EMC

In practice, any conductor can be includedusefully in the equipotential network of themasses: metallic tubes, pipes, drains, cable-trays and ladders, structural members(girders, beams, reinforcing rods, etc.). Sucha mesh of interconnected metal improvesconsiderably the EMC of systems, as well asfavouring the measures for safety of persons.The nature of the conductors has littleimportance in equipotentiality. A steelconductor at HF has approximately the sameinductance as a copper conductor ofequivalent cross-sectional-area and length.These connections to any and all masses inbuildings are authorized and desirable.It is simpler, and ensures the best results, tointerconnect all masses of every kindroutinely, rather than to limit theinterconnections to the masses of theequipments and devices, of the electrical andelectronic systems only. In this way a "meshof masses" or a "mass grid" (both terms areused) is formed. It is rarely useful to install anelectrical conductor; it is sufficient to simplyinterconnect, at as many points as possible,all metallic pipes, tubes, ducts, structuralgirders, beams, reinforcing rods, etc. It isrecommended to connect every large rack,frame, or structure, to the mass grid atintervals of approximately 1 metre.In conclusion, an effective equipotentialcondition of all masses favours thesatisfactory performance of any electronicsystem, especially for rapid or highly-pollutingdigital systems. Whether it be for theimprovement in immunity fromelectromagnetic interference, or for thereduction of radiation from the equipments ofthe installation, the mesh connection ofmasses affords a simple, relativelyinexpensive and efficient solution forfrequencies up to several tens of megahertz.If the public power-supply system benefitsfrom 3-phase star-connected operation, it isbecause the supplies are (and must remain)galvanically isolated each one from theothers. It should be understood that factorswhich are favourable for phase conductors,are not necessarily so for the masses.A HF current cannot flow easily through aphase conductor: it is only possible at lowfrequencies. To divide the inevitable HFcommon-mode currents through the multiplepaths of numerous conductors of a mass gridis a guarantee of protection for the signalcables. Experience shows that when asystem functions correctly in the absence ofHF interference, regardless of the cablingphilosophy of its masses, the routine meshingof masses has no adverse effect on thecorrect operation of the system; on thecontrary, it often decisively improves thesystem performance.


3.2 masses (continued)Except for the installation of a (very costly)"Faraday cage" type of enclosure, the meshconnecting of masses is the only practicalway to guarantee a sufficient level ofequipotentiality to counter effectively all typesof electromagnetic interference.The notion of equipotentiality is more andmore localised as the frequency increases.The equipotential condition is only obtained atHF by the free circulation of the common-mode currents in all directions, i.e. bydispersion.Thus, in an industrial environment, it isrecommended to connect routinely allconducting structures to neighbouringconducting parts of the building by theshortest possible "bonding" conductors and,where appropriate (e.g. in multi-storeybuildings) in three dimensions. This is themost economic way to improve theequipotentiality of an installation at allfrequencies, despite some inevitable currentsin the masses.Only desk-top equipments in an officeenvironment and not connected to a networkhave no need be connected to a mass grid.They must, on the other hand, be verycarefully screened.

Appendix EMC - 15

EMC

lower strength than that of the original field.A cable in close proximity to conductive massfrom end to end is therefore less exposed tothe most severe type of interference, viz: thatof common mode.Attenuation effects can be made moreeffective by arranging the mass, wherepossible, to envelop the conductors to beprotected. In this way, a woven metallicscreen, incorporated in signal cables andconnected to mass, protects the enveloppedconductors above a frequency of 1 MHz withan attenuation factor of at least 300.It is difficult and expensive to shield all theinterconnections in a installation, but it isoften easy to select cable routes whichprovide good attenuation. To benefit from theattenuation effect, it is sufficient to fix cableson conductive mass throughout the entirecable length. Such masses must be carefullybonded together electrically and to all nearbystructural framework. The quality (i.e. the lowimpedance) of interconnecting bonds is ofprimary importance. The most efficient is adirect contact of sheet metal on sheet metal.

3.3 attenuation effects

The attenuating effect of a conductingstructure (mass) is defined by the amplitudeof the common-mode interference appearingon a cable installed at a location remote fromany masses, with respect to the amplitude ofthe interference on the same cable due to thesame disturbance, but with the cable installedin close proximity (i.e. clamped firmly) tomass, throughout its length.

fig. EMC-17.

Electrical continuity from one end to the other,and the correct connection to mass atextremities, guarantee an effectiveattenuation factor. It is recommended toconnect cable ways to conductive buildingstructures at intervals along the cable route.The attenuation factor is not reduced bythese additional contacts between masses,but the mass mesh is improved. In a singlecable tray, in order to limit "cross-talk", powercables or, for example, cables of speedcontrollers should not be placed beside small-signal cables.The ideal, in an industrial environment, wouldbe to install three separate cable trays, i.e.one for measurements and similar functions,one for control and indication circuits, andone for power cables.A copper conductor provides an attenuationfactor of the order 5 if it is installed throughoutthe entire length close to the protected signalcable. It is therefore an advantage toassociate signal cables with interconnectingearthing cables in a common cable way (forinstance, between two buildings). This is stilltrue even if the earths are interconnectedelsewhere. It is always possible to add amass cable adjacent to a particularlysensitive signal cable if necessary. The masscable is then referred to as a "cable ofaccompaniment".A buried cable which is passing an a.c.current in common mode creates a magneticfield in the surrounding soil. This (concentric)field gives rise to (Foucault) currents in theearth and the magnetic energy is dissipatedin the form of heat. The common-mode

attenuation effect of a perforated steel sheet metal, type "dalle marine"direct contact, sheet metal on sheet metal

frequency(MHz)cable

metal flexible-connection tresses

(same connection at 2 extremities)

0,1 0,3 1 3 10 30 100

10

20

40

dB

fig. EMC-16: example of attenuating

effect (in this case equal to 5).

The attenuating effect is one of the keyfactors in EMC, being effective and not toocostly. In order to exchange signals in goodconditions, i.e. in limiting the interferencepicked up by the signal cables, it is importantto reduce common-mode coupling.Any metallic structure, close to, i.e. in contactwith, and longitudinally parallel to a signalcable, from one end to the other, can providetwo favourable effects:1 - A more effective meshing of the masses.

For d.c. currents, the mesh does not act asan attenuator; its role is to reduce theresistance between masses, not to provide ashielding effect. The galvanic effect of themesh is independent of the proximity of thesignal cables with the mass.2 - An attenuation (shielding) effect. Theeffect of proximity adds to that mentioned inabove, if the word "impedance" replaces"resistance". It is achieved by connectingequipments, which are interconnected to themass of conducting structures which areclose to the signal cables. The benefit is anefficient shielding which is practically cost-free. The attenuation effect being directlyattributed to mutual induction, there is noattenuation of d.c. interference, as noted in 1.It should be borne in mind that any cable ispotentially an excellent wide-band antenna,especially in the metric range. In order toreduce its radiation ability, a simple, efficientand inexpensive method consists in placingthe cable as close as possible to a massstructure throughout its length, i.e. close to amass cable, metal ducting, structural girder,etc. The attenuation effect produced by amass conductor close to a signal cable issimply explained, as follows. On the occurrenceof an electromagnetic-wave disturbance, acurrent is induced in the mass conductor.This current generates, according to Lenz’slaw, a magnetic field which acts in theopposite sense to the field that produced thecurrent. A signal cable close to the massconductor will therefore be affected by thedifference only of the original field and thereactive field of the mass-conductor current.The resulting field affecting the signal cable isknown as the residual field and is evidently of

victim cable

I

mass conductor

10 V

victim cable

I 2 V

16 - Appendix EMC

EMC


3.4. installation and cabling rules

To resolve the majority of EMC problems, it issufficient to respect (rigorously) a fewelementary cabling rules. The firstrequirement is to decide to which group eachcable belongs. The following classes of cablegroups cover most practical installations.

Group n° 1 - Measuring circuits (low-level

analog signals) and supplies to analog

probes. This group is sensitive.

Group n° 2 - Digital circuits. This group isalso sensitive (especially to impulses andbursts). It can also interfere with the circuitsof Group n° 1.Group n° 3 - Control and indication

circuits, including «all-or-nothing» (AON)

relays. This group will interfere with Groupsn° 1 and n° 2.

Group n° 4 - Power-supply cables. Theseare power cables from the public distributionnetwork, or from a private generating source(emergency power supply for example).Currents at this level are switched andchopped (by various power-electronicsequipments, rectifiers, inverters, and so on...).In normal operation these functions generateHF current and voltage components, in andon the supply conductors. Such currents andvoltages constitute a highly-pollutedenvironment for Groups n° 1, n° 2 and n° 3.It is recommended that the cables and wiresof each Group have a distinctive and differentcolour to the other Groups.

Rule n° 1 - The "go" and "return"

conductors of any circuit must always be

placed as closely together as possible.

This general rule applies also to power-supply cables. Do not supply in "star" (i.e.radially) two circuits that are not isolated,which exchange signals.It is necessary, even for the signals of AONrelays with one common conductor, to"accompany" the active conductors with atleast one common conductor per cable or permulti-core cable. For analog or digital signals,the use of two-core cables (or pairedconductors) is the basic minimum precaution.

Rule n° 2 - All internal-circuit

interconnecting conductors, cables, etc.

should be fixed in close contact with

equipotential structures constituting the

electrical mass. This measure ensures thebenefit of interference attenuation previouslydescribed, at practically no cost. Ensure thatunused wires, or cables or free cores are not

currents are damped by this effect, which isnot exactly the same as that of theattenuation described above, but is ratheranalogous to the action of a transformer witha resistive load. This damping action is

particularly effective where the interference isdue to repetitive trains of transient dampedoscillations (i.e. "bursts"). The Foucaultcurrents in the soil increase the degree ofdamping.

3.3 attenuation effects (continued)

printedcircuit

no

_ U +

yes

_ U +

printedcircuit

fig. EMC-18.

allowed to move unduly in an equipment.

Rule n° 3 - It is recommended to use

screened cables for noisy and for

sensitive circuits.

Screening is an effective protection againstHF noise, provided that it is connected tomass at least at each end of the cable. It isquite possible to juxtapose two cables ofdifferent groups, provided that at least one(but preferably both) cable(s) is (are)screened and connected by a flexible wovenmetallic tress to mass at each extremity.Screened cables properly installed areimmune to "cross-talk".

Rule n° 4 - Only conductors of the same

Group can be routed together in a cable,

or in the same bundle.

For flat ribbon-type multi-core cables, theconductors carrying analog signals shouldseparated from those carrying digital data byat least two conductors mass-connected tothe reference voltage of each card. For digitalconductors, connecting one wire out of two,of a flat ribbon-type cable, to the zero voltageat each end, reduces the HF cross-talkbetween lines by a factor of 5-10. Moreover, itis detrimental to use one multi-coreconnecting cable link for different Groups. Inpractice, spacing the cables by approximately30 cm is generally sufficient, even in anisolated environment, to reduce the cross-talkto an acceptable level. Crossing two cablesfrom different Groups provides the lowestpossible mutual coupling if the two cablescross at 90°. This practice should thereforebe carried out routinely.

Rule n° 5 - Any free (i.e. unused)

conductors of Groups n° 2 or n° 4 should

always be connected to the mass of the

chassis at both ends. By this means, theattenuation effect can often reach a factorexceeding 2. These connections to massmust be easily removed to free any coreswhich may be needed at a later date. ForGroup n° 1 (at very low voltage andfrequency) such a connection could be adisadvantage and is not recommended.Noise at the industrial frequency could causeunacceptable interference.Rule n° 6 - The cables of Group n° 4 need

not be screened if they are filtered.

It is generally necessary to filter power-supplycables at the point of entry of an equipment.On the other hand, it is difficult to filter powercables supplying speed-change controllers,especially when the peak current is high.It then becomes necessary to screen thecables by flexible metallic tresses or by acontinuous metal tube connected to the massat each end.The opposite case is also true: a cable whichis well screened does not need filtering. In acommon plinth, a screened signal cable haspractically no problems of interference fromneighbouring power-supply cables.

Appendix EMC - 17

EMC

transformer


PE conductor

earth electrode

mainearthbar

mass grid (equipotential mesh)

L1L2L3 N

Since the equipments receive their powersupplies individually and are isolated onefrom the other, supplying different equipmentsby separate power lines is a prudentprecaution. In any case, it is advantageous

fig. EMC-20.

3.5 EMC components and solutions

The EMC of systems requires the use ofspecific components. The following is ananalysis of the conditions of use and theperformance of such specific components,viz: electromagnetic screens, filters,overvoltage limiters, and screened cables.

3.5.1 electromagnetic screens

An electromagnetic screen serves to isolatetwo regions of space: one requiring to beshielded from sources of electromagneticradiations originating in the other.Electromagnetic shielding is alwayscomposed of a conducting envelope, which isgenerally metallic.At low frequencies, the fields E and H arerarely coupled. Shielding against the field E isalways effective, even a conductive ink issufficient. A field H however is very difficult toshield against at LF. It requires the use ofmaterials of very high magnetic permeabilityµ (soft iron, mumetal) and/or low-resistantmetals (copper or aluminium). In any case,magnetic screens have to be thick to beeffective. For d.c. systems, protection can beafforded only by using magnetic materials.The screen must be placed as close to theitem to be protected as possible, and thethickness of the screen must be augmented,the greater its volume. It is for shieldingagainst HF disturbances that screens aremost widely used.All microcomputers, and from now on allvideo games, are fitted with EMC shielding.The role of the screen is to limit the radiationfrom the digital circuits to the antennae of theradio receivers in the neighbourhood.A screen being a passive linear bilateral

device, its action is perfectly reciprocal:its efficiency at a given frequency is identicalwhether protecting its inner space fromexternal radiations or vice-versa.The primary action of an EM screen is that ofa mirror by reflecting EM energy backtowards its source. It is then, the reflection

phenomenon. The part of the EM energy notreflected (no reflector is perfect) ispropagated in the screening material, where itis dissipated as heat : this phenomenon iscalled absorption. If one, or other (or both) ofthese two features is (are) good, then thescreen is fulfilling its purpose.A HF screen must be an efficient conductor(low resistance) but must, above all, presentnegligible leakage. «Leakage» used in thiscontext refers to the penetration through thescreen of radiated EM energy. A leak can bethought of as a small crack in the screen.The higher the frequency of the radiation, theshorter the wave-length, and the smaller thesize of cracks which can be tolerated.Contrary to a widely-held belief, the nature ofthe materials used for shielding is of littleimportance at HF. The only feature whichmust be of the highest quality concerns thelow resistance and electrical contacts: greatcare must be taken to avoid oxydation orother types of corrosion. For this reason,contacts are normally either tin- or nickel-

fig. EMC-19.

transformer

noisyplant

sensitiveequipment

to be avoided ! better ! excellent !

∆ ∆ ∆ ∆

Rule n° 7 - Noisy equipments should be

supplied by separate power cables.

This rule will minimize the supply-systemnoise in differential mode. The rule should notbe confused with the practice of radialconnections of the masses previouslydiscussed. The neutral conductor should notbe connected to the mass, except at a singlepoint. This is the fundamental differencebetween a neutral conductor and a PE(Protective Earth) conductor*.* The TN-C scheme for power installations uses oneconductor only for both PE and neutral functions. Forobvious reasons, the TN-C scheme is never used whereEMC is important.

that the masses (chassis) of the equipmentsare all maintained at the same potential:radial network for power supplies, mesh-connected network for masses.The connection of the main earth bar at themain distribution board of an installation (seebelow) to the mass grid should preferablyhave an inductance of less than 1 micro-Henry (the lower the better). A singleconductor of 50 cm or two parallel conductors(not too close) of 1 m each, etc.

Power supply and connections to masses

of an electrical equipment

For supplies to an installation, it is anadvantage to locate the transformer asclosely as possible to the items of load,while taking account of the effect of staticmagnetic-induction fields.

18 - Appendix EMC

EMC

3.5.2 EMC filters

An EMC filter is a protection againstinterference by conduction and is generallymade up of a combination of capacitors andinductors. Its role is to allow the passage ofenergy or signals within the band of usefulfrequencies and to reject parasiticfrequencies.Filters in the power-supply circuits are all low-pass filters which allow power-frequencycurrents to flow, but which suppress currentsof higher frequencies. For an interconnectingcoaxial cable, a high-pass filter is anti-parasitic: it allows the HF signal to pass, butrejects any LF interference current.The cable can then be connected to mass atboth ends without any difficulty. A filter at theinput to a radio receiver is a band-pass filterwhich rejects signals outside the band offrequencies required (as well as anyinterferences). Finally, a harmonic filter is anotch filter, which is tuned to act as a short-circuit (generally phase/phase) at a harmonicfrequency, usually two or more filters for thefirst several odd-numbered harmonics abovethe fundamental frequency, since theseinvariably have the greatest magnitude.An EMC filter being a linear circuit as long asthe inductors remain unsaturated) andpassive, is also bilateral. It is equally effectiveat a given frequency from the interior to theexterior, as in the opposite sense. A filterfunctions firstly by reflection, i.e. by sendingthe energy back towards the source, due to amismatch of the filter/line impedances, thenby absorption, i.e. loss of energy in the formof heat, as it passes through the filter.Since inductors are low-loss components atLF, the L-C filters function principally byreflection. The effectiveness of a filter alsodepends on the upstream and downstreamimpedances. If these impedances vary, theefficiency of the filter referred to as "insertionlosses", will vary. Remark: if a filter canmismatch a line, there is the possibility that itmay also match a line. This is a phenemenonwhich may be observed on LF power-supplyline filters: a resonance (even partial) of thefilter results in a deterioration, at LF, of thelevel transmitted, compared to that when nofilter is present.It should be verified that the resonantfrequency of the filter is not likely to be aproblem (it should be below that of thecurrent-chopping frequency, for example).Filters in power-supply circuits usually useinductors in common mode, also called"current-compensation coils" or"compensated inductors".


3.5 EMC components and solutions (continued)

Such filters evidently present differentdegrees of effectiveness in common modethan in differential mode.If the inductor of a filter is saturated by thecurrent flowing through it, the effectiveness ofthe filter is greatly impaired.In order to respect the EMC standards, a filteris practically obligatory in power-supplycircuits. Where no filter has been installed, itis often necessary to select one having anefficacity in the order of 30 dB in commonmode at 100 MHz. A power-supply filter must,in order to perform efficiently at HF, beinstalled according to three rules:1 - Screw the filter sheet-metal to sheet-metalin order to limit its impedance to the mass.2 - Arrange the supply cable to enter the filterat the opposite face to that of the outputcircuit, in order to limit common-modeupstream/downstream coupling.3 - Fix the cables firmly (i.e. clamped) againstthe sheet-metal of the unit to limit radiationfrom the upstream conductors affecting thedownstream circuit.Preferred practice is to install all filters of anequipment on the same metal base whichserves as the potential reference. The notionof equipotentiality at HF is local: eachequipment has to provide, by means of itsconducting envelope, its own potentialreference to the input and output filters and toshielded connection cables.The signal filters are often R-C combinations.A simple resistor of the order of 1 kΩ in serieswith a sensitive line can suffice to reinforce itsimmunity. Small inductors in common modecan also be used, with 2, 4, or more, wireswound "two wires in hand".It is interesting to note that these componentsreduce common-mode interference, withoutaffecting the useful signals transmitted indifferential mode.

plated. An electromagnetic shield need notnecessarily be earthed to be effective. For amagnetic field simply its presence issufficient. For electric fields, it is enough thatthe screen acts as potential reference for theinput and output circuits. It may be concludedthat a shield prevents the fields frompenetrating the protected space, but also,and especially, prevents parasitic currentsfrom entering. Thus, shields and filters arenot rivals, but are complementary, one withthe other.If a screen is of excellent quality, with noleakage, it is possible to install input and

output connectors at any convenient point. If,on the contrary, a screen performs badly, withexcessive leakage (display, keyboard,printed-circuit board or disk reader) then, it isan advantage to group all the input andoutput connectors on a common chassis,remote from the leakage, the role of thischassis being that of a reference-potentialpoint. It may be noted that all modernmicrocomputers have their cables grouped atthe rear face, remote from the disk unitswhich are mounted on the front face.

Appendix EMC - 19

EMC3.5.3 overvoltage protection

The role of an overvoltage limiter, sometimesreferred to as "surge diverter" or "lightningarrester" (depending on its intended location)is to reduce the risk of destruction tocomponents or entire equipments byinterferences which may occur at excessivevoltage levels.An overvoltage limiter is generally a non-linear unilateral device: it limits the peak valueof voltage at a level which is much lower thanthat of the incoming surge. This reduced levelbeing, in principle, lower than the ratedimpulse withstand capability of alldownstream plant and equipment.The limiting of the voltage peak, however,does not reduce the HF radiation fieldstrength. Conversely, a low-pass filter doesnot limit the voltage peak, the duration ofwhich, at half-peak level, considerablyexceeds the response time of the filter.Thus, an efficient filter which suppressesfrequencies above 10 kHz would have a rise-time of about 35 µs. This filter cannot limit anovervoltage due to lightning, the tail-time ofwhich to half-peak is standardized at 50 µs.The first voltage limiters used in telephonesystems were gas-filled discharge devices.A gas-filled glass envelope contains twoelectrodes separated one from the other by acalibrated space. An overvoltage ionizes thegas which allows a discharge to occurbetween the electrodes, thereby reducing thepotential and allowing the gas to de-ionize.Such a component is robust and has only asmall parasitic effect.Its occasional failure, often by short-circuitingof the electrodes (i.e. following anovervoltage discharge, the ionized gassometimes provokes a short-circuit at normalworking voltage) means that its reliabilitycannot be guaranteed. In order to protect a

public service supply line, the low arc-voltage,several tens of volts, requires the installationof a varistance in series, to extinguish the arcwhen the surge has been dissipated.Analogous components exist at high voltage("horned spark-gaps" for example). At lowvoltage, "silicon spark-gaps" such as "Trisil"of Thomson (a triac controlled by a Zenerdiode in the trigger circuit) are well adapted tothe protection of telecommunication lines andcircuits.The highly non-linear metal-oxide varistancecomponents are well adapted to theprotection of supply-circuits. A disk of zinc-oxide becomes conductive when the voltageapplied to its two faces exceeds a"knee-point" value. That voltage, proportionalto the thickness of the disk, varies from sometens of volts to several kilovolts. The energythat a component can tolerate depends onthe volume of the disk: from tens of Joules tosome tens of thousand Joules. The maindrawback of varistances is their degradationduring periods of conduction.Zener diodes of very low dynamic resistancehave a precise knee-point voltage and a shortresponse time. Their low energy handlingcapacity, of a fraction of a Joule to severalJoules, limits the use of these components tothe protection of signal circuits. Failure ofsuch a diode always occurs as a short-circuit,a condition that guarantees "fail-safe"protection for the circuits.In all cases, an overvoltage device incommon mode should be connected directlyat the mass of the item to be protected, andnot, as is still often the case, by a long cableconnected radially from a distant earth bar.The response time of an overvoltage limitingdevice depends on the length of itsconnections.

20 - Appendix EMC

EMCLocal networks present at least one particularproblem: the numerous equipments arespread out, relatively distant one fromanother, and are installed for userconvenience rather than for EMCcoordination considerations, are oftensupplied from different lines, andinterconnected by conventional wiringpractices.

4. local network problems

Such an installation invariably creates anumber of mass loops of very large area.The interconnection of equipments createslarge mass loops. One of the most seriousdangers for local networks is the magneticfield in the areas of the mass loops, createdby the current from a lightning stroke. It maybe noted that a surge induced in the interiorof a building, on average once a year, givesrise to an overvoltage which can attain, orexceed, 100 volts per square metre of looparea.The meshing of masses should be carried outin the three dimensions (laterally andvertically), especially in multistorey buildingswith network equipments on several floors.Two adjacent floors must be meshed togetherby all conducting metal work which passesthrough the intervening floor. Themultiplication of these conductors affords thefollowing advantages:1 - Improvement of the "verticalequipotentiality" of the building by effectivelyreducing the value of loop inductances, andconnecting them in parallel.2 - Improvement of the "horizontalequipotentiality" of the building and thesymmetrical flow of surge current directly toearth.3 - Reduction of induction from the magneticfield in the interior of the building. At a pointmidway between two parallel conductorspassing equal currents in the same direction,the magnetic field strength H is zero.Experience has shown that if the masses areineffectively meshed, and interconnectingcables are without attenuation effects, someprinted-circuit boards can be expected to bedestroyed by the induction of even a distantlightning stroke. On the other hand, if themasses are reasonably well interconnected,with conducting cable trays screwed firmly tothe equipment metal frames, a lightningstroke (even direct) produces minorinterference, and causes no destruction ofelectronic circuits. In a badly-meshedenvironment, only equipments completelydisconnected or well shielded are out ofdanger from lightning.

signal cable

I

field H

peripheralcontrol unit

supply cable

supplycable

fig. EMC-21: interconnection of

equipment forms loops with earthed

conductors.

The current induced in a mass loop by themagnetic field of a lightning stroke has thesame form as that of the stroke current; it canexceed 100 A in the case of a large loop.The best solution for limiting the risks ofdestruction is to bring together, for example ina common tray, the signal cables and thepower-supply cables. A shielded cable, withits screen properly connected to mass ateach end is free from cross-talk interference.The presence of a mass cable in intimate(e.g. clamped) contact with a cable (signal orpower-supply) reduces typically by a factor of3 or 4 the interference caused by lightning,provided that the mass cable is connected tomass (at least) at each end. A bolted metalcable duct throughout the entire route of acable has an attenuation factor of the order30.The woven metal screen of a shielded cable,with short, direct connections to mass at bothends, reduces the induced voltage by a factorof about 100.Local networks processing large amounts ofdata require that the characteristic impedanceof interconnecting signal cables be matchedto the input/output impedances of theinterconnected equipments, in order to avoidlosses by reflections due to mismatchedimpedances.If one of the two matching units of a long lineis disconnected, transmission becomesimpossible.A frequent problem on local networks, apartfrom software parameters, is the loss ofavailability caused by electromagneticinterference. The software "filters" errors, butthe useful output is sometimes severelyreduced. The user only notices theseproblems in the rare cases of permanentinterference. The simple observance of theserveral EMC installation rules cited in theforegoing text is sufficient to resolve theseproblems.

fig. EMC-22: lightning produces

interference more often by induction than

by a direct-stroke current.

insulationstressed by 15 kV

signal cable

∆H∆t

aera ≈ 300 m2

∆I/∆t = 100 kA/µs

400 m



Appendix EMC - 21

EMC

list of "cahiers techniques"

N° CT English French Spanish114 Residual current devices X X X141 Les perturbations électriques en BT X X144 Introduction to dependability design X X145 Etude thermique des tableaux électriques BT X X147 Initiation aux réseaux de communication X

numérique148 High availability electrical power distribution X X149 EMC: Electromagnetic Compatibility X X150 Development of LV circuit breakers X X X

to standard IEC 947-2152 Harmonics in industrial networks X X X154 LV circuit breaker breaking techniques X X X155 MV public distribution networks X X X

throughout the world156 Sûreté de fonctionnement et tableaux électriques BT X X158 Calculation of short-circuit currents X X X159 Inverters and harmonics X X X

(case studies of non-linear loads)160 Harmonics upstream of rectifiers in UPS X X161 Automatic transfering of power supplies X X

in HV and LV networks162 Les efforts électrodynamiques X X

sur les jeux de barres en BT163 LV breaking by current limitation X X166 Enclosures and degrees of protection X X X167 Energy-based discrimination X X X

for low-voltage protective devices172 Earthing systems in LV X X X173 Earthing systems wolrdwide and evolutions X X X179 Surtensions et parafoudres en BT, X

Coordination de l'isolement en BT

Documents

0100 Electrical Installation Guide Guide En