Cahier technique no. 114 - pangonilo.com

..........................................................................

Cahier technique no. 114

Residual current devices in LV

J. Schonek

Collection Technique

Building a New Electric World

«Cahiers Techniques» is a collection of documents intended for engineersand technicians, people in the industry who are looking for more in-depthinformation in order to complement that given in product catalogues.

Furthermore, these «Cahiers Techniques» are often considered as helpful«tools» for training courses.They provide knowledge on new technical and technological developmentsin the electrotechnical field and electronics. They also provide betterunderstanding of various phenomena observed in electrical installations,systems and equipments. Each «Cahier Technique» provides an in-depth studyof a precise subject in the fields of electrical networks, protection devices,monitoring and control and industrial automation systems.

This constantly updated collection can be downloaded from:http://www.technical-publications.schneider-electric.com

Please contact your Schneider Electric representative if you want either a"Cahier Technique" or the list of available titles.

The "Cahiers Techniques" collection is part of the Schneider Electric’s "Col-lection technique".

Foreword

The author disclaims all responsibility subsequent to incorrect use of infor-mation or diagrams reproduced in this document, and cannot be heldresponsible for any errors or oversights, or for the consequences of usinginformation and diagrams contained in this document.Reproduction of all or part of a "Cahier Technique" is authorised with thecompulsory mention:"Extracted from Schneider Electric "Cahier Technique" no. ..." (please specify).

no. 114Residual current devices in LV

ECT 114 first issue, february 2006

Jacques Schonek

A graduate of the ENSEEIHT engineering school with a doctorate inengineering from the University of Toulouse, J. Schonek was involvedin the design of Telemecanique variable-speed drives from 1980 to1995.He then went on to manage the harmonic-filtering activity.He is currently working as an expert in Electrical DistributionApplications in the Architectures and Systems group of SchneiderElectric.

Cahier Technique Schneider Electric no. 114 / p.2

Lexicon

Common mode disturbanceAny continuous or transient electromagneticphenomenon occurring between a live part of apower system and earth. May be a transientovervoltage, a continuous voltage, anovercurrent or an electrostatic discharge.

Differential mode disturbanceAny phenomenon between different live parts ofa power system, e.g. an overvoltage.

Direct contactContact of a person with the live parts ofelectrical devices (normally energised parts andconductors).

Earth-leakage currentCurrent that flows from the live parts to earth, inthe absence of an insulation fault.

ElectrisationApplication of voltage between two parts of thebody of a living being.

ElectrocutionElectrisation resulting in death.

Exposed conductive part (ECP)Conductive part likely to be touched and which,although normally insulated from live parts, maybe energised up to a dangerous voltage leveldue to an insulation fault.

Fault current IdCurrent resulting from an insulation fault.

Indirect contactContact of a person with accidentally energisedexposed conductive parts (ECP), usually due toan insulation fault.

InsulationArrangement preventing transmission of voltage(and current flow) between a normally energisedpart and an exposed conductive part (ECP) orearth.

Insulation faultBreak in insulation causing an earth-fault currentor a short-circuit via the protective conductor.

Intentional leakage currentCurrent that flows to earth via the intentionallyinstalled components (resistors or capacitors), inthe absence of an insulation fault.

Live conductorsSet of conductors for electrical powertransmission, including the neutral, with theexception of the PEN conductor for which the“protective conductor” (PE) function takes priorityover the “neutral” function.

Natural leakage currentCurrent that flows to earth via the insulation, inthe absence of an insulation fault.

Neutral systemSee "System earthing arrangement".

Protective conductors (PE or PEN)Conductors which, according to specifications,connect the exposed conductive parts (ECP) ofelectrical equipment and certain other conductiveparts to the earth electrode.

Rated residual operating current I∆∆∆∆∆nValue of the residual operating current assignedby the device manufacturer at which the devicemust operate under the specified conditions.According to construction standards, at 20 °C,low voltage residual current devices mustoperate at residual currents between I∆n/2 andI∆n

Residual currentAlgebraic sum of the instantaneous values of thecurrents flowing through all live conductors in acircuit at a point of the electrical installation.

Residual current device (RCD)Device whose decisive quantity is the residualcurrent. It is normally associated with orincorporated in a breaking device.

Residual operating currentValue of the residual current causing a residualcurrent device to operate.

System earthing arrangement (SEA)Also referred to as the neutral system, earthingsystem or earthing arrangement.Standard IEC 60364 stipulates three main typesof earthing arrangements that define the possibleconnections of the source neutral to earth and ofthe exposed conductive parts (ECP) to earth orthe neutral. The electrical protection devices arethen defined for each one.

Touch voltage limit (UL)Voltage UL below which there is no risk ofelectrocution.

Ventricular fibrillationA malfunctioning of the heart corresponding toloss of synchronism of the activity of its walls(diastole and systole). The flow of AC currentthrough the body may be responsible for this dueto the periodic excitation that it generates. Theultimate consequence is stoppage of blood flow.


Residual current devices in LV

Contents

1 Introduction p. 4

2 The risks of electrical currents 2.1 Electrisation of persons p. 4

2.2 Fire hazards p. 6

2.3 Damage to equipment p. 7

3 Protection against the risks 3.1 Installation rules p. 8of electrical currents 3.2 Detection of insulation faults p. 9

4 RCD operating principle and description 4.1 Operating principle p. 11

4.2 Applications p. 11

4.3 Main characteristics p. 12

4.4 Technology p. 13

4.5 Constraints due to the current sensor p. 15

4.6 Special applications p. 17

5 Conclusion p. 22

Appendix 1: Calculation of touch voltages p. 23

Appendix 2: Types of converters and fault-current waveforms p. 25

Appendix 3: Leakage currents for different system earthing arrangements p. 28

Appendix 4: RCD thresholds and power system voltages p. 30

Bibliography p. 31

Today, residual current devices (RCD) are recognised as the most effectivemeans of protecting life and property against electrical hazards in low-voltage systems.

Their selection and optimum use require sound knowledge of the principlesand rules governing electrical installations and in particular system earthingarrangements as well as existing technologies and their performancelevels.

All these aspects are dealt with in this "Cahier Technique", with in additionnumerous answers provided by Schneider Electric technical andmaintenance departments to frequently asked questions.


1 Introduction

Compared to other energy sources, electricityhas many advantages, but also many risks. It isused on a daily basis by the general public andmany accidents still occur, resulting in burns,fires and electrocution.

Strict installation rules have been set up byinternational (IEC, CENELEC) and national(e.g. NFPA in the USA and UTE in France,)organisations.

Dependable protective devices have beendesigned by carefully analysing the risks andconsequences of equipment failures or incorrectuse. Among these devices, RCDs (residual

2 The risks of electrical currents

2.1 Electrisation of persons

A person subjected to an electrical voltage iselectrised. Depending on the level ofelectrisation, the person may be subjected todifferent pathophysiological effects:c disagreeable sensation,c involuntary muscular contraction,c burns,c cardiac arrest (electrocution).

These effects depend on various factors,including the physiological characteristics of theperson, the environment (e.g. wet or dryconditions) and the characteristics of the currentflowing through the body.

A person may be subjected to an electricalshock in two manners:c direct contact, e.g. the person touches anenergised, bare conductor,c indirect contact, e.g. the person touches ametal part of an electrical machine or device withan insulation fault.

The dangerous aspect is the current (magnitudeand duration) flowing through the human bodyand particularly near the heart.

Uc

Phases

3

Id Uc

Fig. 1 : Direct and indirect contact.

a) Direct contact

b) Indirect contact

current devices) are recognised by internationalstandardisation organisations as an effectivemeans to protect life and property.

This document will present the subject in threesteps:c a description of the risks related to electricalcurrents,c an overview of the protection techniquesemployed to limit those risks,c an in-depth presentation of how RCDsoperate.


Figure 2 sums up the work of the InternationalElectrotechnical Commission on the subject(standard IEC 60479-1, Ed.4, 2005, Effects ofcurrent on human beings and livestock - Part 1.General aspects). It indicates the consequencesof AC current flowing through the human body,from the left hand to the feet, depending on thecurrent and its duration.

It is especially important to consider zones AC-3and AC-4 where there is real danger.

c Zone AC-3 (between curves B and C1)

Usually no organic damage, but there is alikelihood of muscular contractions and difficultyin breathing, with reversible disturbances in theformation and conduction of impulses in theheart.

c Zone AC-4 (located to the right of curve C1)In addition to the effects noted for zone AC-3,

the probability of ventricular fibrillation:

v increases up to about 5% between curves C1and C2,

v increases up to about 50% between curves C2and C3,

v exceeds 50% beyond curve C3.

The probability of dangerous pathophysiologicaleffects such as cardiac arrest, breathing arrestand severe burns increases with current and time.

Note that a 150 mA current may flow in a personin contact with a 230 V voltage, underunfavourable conditions.

Given the current levels considereddangerous, a maximum permissible value of30 mA is considered safe.

For LV systems, the dominant component inbody impedance is the skin resistance, which

AC-1 : PerceptionAC-2 : Involuntary muscular contractionsAC-3 : Difficulty in breathingAC-4 : Serious pathophysiological effects

AC-4.1 : probability of ventricular fibrillation increasing up to about 5 %AC-4.2 : probability of ventricular fibrillation up to about 50 %AC-4.3 : probability of ventricular fibrillation above 50 %

Fig. 2 : Time/current zones of effects of AC currents (15 Hz to 100 Hz) on persons.

Body currentIs (mA)

10

20

50

100

200

500

1 000

5 000

10 000

2 000

C1 C2 C3

Duration of current flow I (ms)

A B

AC-2 AC-3 AC-4

0.1 0.2 0.5 1 2 5 10 20

Threshold = 30 mA

50 100 200 5001 000

2 0005 000

10 000

AC-1

AC-4.2

AC-4.3

AC-4.1


depends essentially on the environment (dry,humid or wet conditions).

IEC has defined the "conventional touch voltagelimit", noted UL, as the maximum touch voltagethat can be maintained indefinitely under thespecified environment conditions. The valueused is 50 V AC rms. This value is consistentwith an average impedance value of 1700 Ω anda maximum current of 30 mA.

Effects as a function of voltage and frequency

IEC 60479-1 provides curves showing thevariation in body impedance depending on thevoltage and the frequency.

Figure 3 shows that body impedance decreaseswith frequency. However, note that IEC 60479-2(Effects of current on human beings andlivestock - Special aspects), dealing with theeffects of AC current at frequencies above100 Hz, indicates that the threshold current forventricular fibrillation at 1000 Hz is approximately14 times greater than at 50/60 Hz current

2.2 Fire hazards

A study carried out in the 1980s and 1990s inGermany by an insurance company on fires onindustrial and commercial premises revealed thatelectricity was the cause of over 40 % of the fires.

The cause of many electrical fires is a majorshort-duration temperature rise or electric arcdue to an insulation fault. The risk increases withthe level of the fault current. It also depends on

the level of fire or explosion hazard specific tothe room (storage of flammable materials,presence of volatile hydrocarbons, etc.).

Many electrical fires are caused by acombination of factors:c an old installation,c wear of insulation,c accumulation of dust and humidity.

6 000

10 V CA

25 V

50 V

100 V

225 V

1 000 V

5 000

4 000

5 000

2 000

1 000

600

0

50 100 250 500 1 000 2 000

775 V

Frequency (Hz)

Total body impedance

Fig. 3 : Total body impedance ZT as a function of the frequency and the touch voltage.


The progressive increase in tracking currents onthe surface of polluted and damp insulationresults in small discharges that cause carbondeposits. This phenomenon is related to surfacecondensation and drying cycles and thereforeevolves very slowly. If the tracking currentexceeds 300 mA, an avalanche phenomenon

occurs that can inflame the carbon depositswhich, in turn, may inflame the insulation anddevices.A 300 mA leakage current represents a real firehazard. The leakage current flows from thesource to the ECPs, but does not return to thesource via the return conductor.

Fig. 4 : Origin of fires in buildings.

Fig. 5 : Process resulting in a fire.

Bare flamme 37% Lightning 1%

Explosion 1%

Electricity 41%

Other 7%

Cigarette 6%

Accident 7%

Leakagecurrent

Small discharges

Carbonised insulation and dust

2.3 Damage to equipment

temperature rise in conductors and may lead toinsulation breakdown and the flow of an earthfault current. The latter may remain at a lowlevel, difficult to detect, or rapidly degenerateinto a short-circuit and provoke major damage.

Certain electrical equipment may be damaged ordestroyed by high currents. This is the case formotors during prolonged operation above ratedload and cables if too many devices areconnected. The overcurrent provokes excessive


3 Protection against the risks of electrical currents

3.1 Installation rules

The international reference standard is IEC 60364,Electrical installations in buildings, and in particularpart 4-41, Protection for safety - Protection againstelectric shock. It lays down installation rules suchthat dangerous live parts are not accessible andaccessible conductive parts are not dangerousunder normal and fault conditions.The standard has been adopted by many countries,as is or with local adaptations. In France forexample, low-voltage electrical installations mustcomply with standard NF C 15-100.

General rulesInstallations must be designed to provide generalprotection against direct contacts during normaloperation and additional protection against indirectcontacts in the event of a fault.

c General protection is implemented by isolating thelive parts, using barriers and enclosures.Protection in the even of faults is implemented byone or more of the following means:v automatic disconnection of the supply,v double or reinforced insulation,v electrical separation (use of an isolatingtransformer),v use of very low voltage.

Automatic disconnection of supply is the mostcommon solution. It entails certain requirements:v earthing of ECPs and equipotential bonding,v specification of a maximum disconnection time fora fault (e.g. 0.4 seconds for 230 V).A protective device must automatically isolate acircuit or device from its supply so that, following afault between a live part and an ECP of a circuit ordevice, a touch voltage greater than theconventional limit cannot be present for a timesufficient to create a danger for a person in contactwith simultaneously accessible conductive parts.c Additional protection is required in case theprotection against direct contacts fails. In particular,IEC 60364-4-41 requires installation of a device forprotection against direct contacts for AC circuitssupplying general-usage socket-outlets up to 20 Alocated outdoors and for socket-outlets up to 32 Aused to supply portable equipment intended foroutdoor use . The operating threshold of the devicemust be 30 mA.

System earthing arrangements (SEA)Standard IEC 60364 defines three main systemearthing arrangements (SEA, see Fig. 6), usedin different manners depending on the country.See Cahiers Techniques publications no. 172and 173.

They differ according to whether or not theneutral point of the voltage source is earthed andthe manner in which the ECPs are earthed. SEAselection depends on installation characteristicsand on operating conditions and requirements(environment, monitoring devices, continuity ofservice).

c TT systemIn the TT system:v the source neutral is connected to an earthelectrode separate from that of the ECPs,v all ECPs must be connected to a singleearthing system dedicated to the installation.

c TN systemThis principle of the TN system is to transform allinsulation faults into a phase-to-neutral short-circuit. In this system:v the LV neutral point of each source is directlyearthed,v all ECPs in the installation are earthed, i.e.connected to the neutral by the PE (withseparate neutral in TN-S systems) or the PEN(combined with the neutral conductor in TN-Csystems) protective conductor.

c IT systemIn the IT system, the transformer neutral is:v isolated from earth (isolated neutral),v or connected to earth via a high impedance(impedant neutral),v all installation ECPs are earthed.


Directy earthed neutral (TT)123N

ECPs connected to neutral (TN-C)123PEN

IMD: Insulation Monitoring Device

ECPs connected to neutral (TN-S)123NPEPE

PE

Isolated neutral (IT)123N

RB RA

RB

RB

RB

IMD

Fig. 6 : The three main SEAs are the TT, IT and TN systems defined by IEC 60364-1. The TN system may beTN-C (neutral and PE combined) or TN-S (separate neutral and PE).

3.2 Detection of insulation faults

An insulation fault may be the consequence ofinsulation deterioration:c between two live conductors,c between a conductor and the ECPs or theprotective conductor,c on a single live conductor, making theconductor accessible to touch.An insulation fault between live conductorsbecomes a short-circuit.In all other cases, a fault (in common mode)causes current to flow to earth. This current,which does not flow back via the live conductors,is called the earth-fault current. It is the algebraicsum of the instantaneous values of the currentsflowing in the live conductors, hence the name"residual current".Remark: If the currents are sinusoidal, Fresnelvector representation may be used and it ispossible to speak of the "vector sum" of thecurrents. However this representation is notrelevant in the presence of harmonic currentsand the term "algebraic sum" is therefore moregenerally applicable.

This current may be due to an insulation faultbetween a live conductor and the ECPs (risk ofindirect contact) or to a failure of the measuresused to insulate or isolate live parts (risk of directcontact). These situations are shown in figure 7next page.

The insulation-fault current between a phaseand earth (common mode) depends on the typeof fault and on the SEA. The current maycreate a dangerous touch voltage, requiringdisconnection of the faulty circuit.See Appendix 1 for a summary of calculationson fault currents and touch voltages dependingon the SEA.

In the TN system, the fault current is equivalentto a short-circuit. The current is high and thecircuit can be disconnected by an overcurrentprotective device.

In the TT system, however, the current is too lowto be detected and cleared by standardovercurrent protective devices (circuit breakerthermal or magnetic protection, fuses).


Fig. 7 : Fault current Id = residual current.

a) Indirect contact b) Direct contact

Ud

N

PE

RB RA

Id

N

PE

RB RA

Id

Similarly, in all cases of direct contact, the faultcurrent is low and cannot be detected andcleared by standard overcurrent protectivedevices. This is also the case for leakagecurrents that constitute fire hazards.

Under these conditions, the fault current must bedetected and cleared by a special device, i.e. aresidual-current device (RCD), discussed in thenext section.


4 RCD operating principle and description

4.1 Operating principle

The operating principle of an RCD is shown infigure 8.A sensor comprising a toroid that surrounds theconductors detects the algebraic sum of thecurrent in the live conductors (phases andneutral).The toroid winding detects variations in the fluxinduced by the residual current.In the absence of an insulation fault, thealgebraic sum of the currents in the conductorsis equal to zero and the toroid does not detectany flux.If an insulation fault occurs, the sum is no longerequal to zero and the fault current in the toroidgenerates a current in the winding.This current is rectified, filtered and amplified.If the resulting signal is greater than a setthreshold, a time delay is initiated (it may beequal to zero for an instantaneous response).If the fault is still present at the end of the timedelay, an opening order is issued for a controldevice.

Use of an RCD is not possible in the TN-Csystem because the neutral and protectiveconductors are not separate and the RCDcannot distinguish between the residual andneutral currents.

Fig. 8 : Operating principle of an RCD.

4.2 Applications

Additional protection against direct contactsAn RCD can detect low leakage currents thatcould flow through the body of a person. It thusprovides additional protection if the normalprotection means fail, e.g. old or damagedinsulation, human error, etc. This can also bereferred to as ultimate protection because it caninterrupt the current even if the other deviceshave failed.Use of a 30 mA RCD on all circuits supplyingsocket-outlets up to 20 A is now mandatory, asper IEC 60364-4-41, Electrical installations inbuildings, Protection for safety - Protectionagainst electric shock.Note that an RCD does not limit theinstantaneous current flowing through the body,but does limit the time the current flows. Notealso that for a direct contact with a 230 V phaseconductor, the flowing current would beapproximately 150 mA. RCDs with 10 or 30 mAsensitivities let the same current through.The two sensitivities provide equivalentprotection. However, the 30 mA threshold

provides a cost-effective compromise betweensafety and continuity of service.Downstream of an RCD, it is possible to supply anumber of loads or circuits as long as theleakage current does not trip the RCD. For agiven leakage current, a reduction in thethreshold makes it necessary to increase thenumber of protective devices.

Protection against indirect contactsAn RCD is the only solution to protect againstindirect contacts on a TT system because thedangerous fault current is too low to be detectedby overcurrent protective devices. It is also asimple solution for the TN-S and IT systems. Forexample, when the supply cable is very long, thelow fault current makes it difficult to set theovercurrent protective devices. And when thelength of the cable is unknown, calculation of thefault current is impossible and use of an RCD isthe only possible solution.Under these conditions, the RCD operatingthreshold must be set to somewhere between afew amperes and a several tens of amperes.

Waveform processing

Threshold

Time delay

Actuator

Σ I ≠ 0


Protection against fire hazardsIEC 60364-4-42 (Electrical installations inbuildings, Protection for safety - Protectionagainst thermal effects) also recognises RCDeffectiveness in protecting against fire hazardsby requiring their use with a maximum operatingthreshold of 500 mA.

This threshold should be reduced to 300 mA inthe near future, as already recommended bycertain national standards such as NF C 15-100in France.

4.3 Main characteristics

An RCD must be selected taking into accountthe type of load supplied. This applies inparticular to semiconductor-based devices forwhich fault currents are not always sinusoidal.Examples of semiconductor converters areprovided in Appendix 3, with the waveforms ofthe fault currents and the corresponding type ofRCD.

Type AC, A, B

Standard IEC 60755 (General requirements forresidual current operated protective devices)defines three types of RCD depending on thecharacteristics of the fault current.c Type ACRCD for which tripping is ensured for residualsinusoidal alternating currents.c Type ARCD for which tripping is ensured:v for residual sinusoidal alternating currents,v for residual pulsating direct currents,v for residual pulsating direct currentssuperimposed by a smooth direct current of0.006 A, with or without phase-angle control,independent of the polarity.c Type BRCD for which tripping is ensured:v as for type A,v for residual sinusoidal currents up to 1000 Hz,v for residual sinusoidal currents superposed bya pure direct current,v for pulsating direct currents superposed by apure direct current,- for residual currents which may result fromrectifying circuits, i.e.:- three pulse star connection or six pulse bridgeconnection,

- two pulse bridge connection line-to-line.with or without phase-angle monitoring,independently of the polarity.

Certain electronic devices can generate faultcurrents not described above. Examples areprovided in Appendix 2. The IEC has begunstudies to cover these special cases as well.

Sensitivity

RCD sensitivity is expressed as the ratedresidual operating current, noted I∆n.Preferred values have been defined by the IEC,thus making it possible to divide RCDs into threegroups according to their I∆n value.

c High sensitivity (HS): 6 – 10 – 30 mA,

c Medium sensitivity (MS): 0.1 – 0.3 – 0.5 – 1 A,

c Low sensitivity (LS): 3 –10 – 30 A.

RCDs for residential or similar applications arealways high or medium sensitivity.

It is clear that High Sensitivity (HS) is most oftenused for direct-contact protection, whereas MSand in particular the 300 and 500 mA ratings areindispensable for fire protection. The othersensitivities (MS and LS) are used for otherneeds such as protection against indirectcontacts (mandatory in the TT system) orprotection of machines.

Break time

As indicated in section 1, the effects of electricalcurrents depend on their magnitude andduration. RCD break times are specified in theproduct standards:

c IEC 61008, Residual current operated circuit-breakers without integral overcurrent protectionfor household and similar uses (RCCBs).


c IEC 61009, Residual current operated circuit-breakers with integral overcurrent protection forhousehold and similar uses (RCBOs).

c IEC 60947-2, Low-voltage switchgear andcontrolgear, Annex B, Circuit-breakersincorporating residual current protection.

c IEC 60947-2, Low-voltage switchgear andcontrolgear, Annex M, Modular residual currentdevices - MRCD (without integral current-breaking device).

The standardised break times are indicated inthe table of figure 9 and in the curves infigure 10 for G and S type devices.

c G (general use) for instantaneous RCDs (i.e.without a time delay)

c S (selective) for RCDs with a short time delay(used in France, for example, for service-connection circuit breakers).

1

t (s)

S

G

0.1

0.04

0.15

0.3

0.010.1 1 2 5 10

Fig. 10 : Maximum break-time curves for S (selective)and for G (general use) RCDs.

Fig. 9 : Standardised values of maximum break times and non-actuating times as per IEC 61008.

4.4 Technology

RCD classification depending on supplymode:"Without auxiliary source" or "Functionallyindependent of line voltage".In this type of device, the tripping energy issupplied by the fault current. This highlydependable supply mode is recommended forresidential or similar applications where the useris not aware of the dangers of electricity. Manycountries, particularly in Europe, recognise the

effectiveness of these devices for residential andsimilar uses (standards EN 61008 and 61009)."Without auxiliary source" or "functionallydependent on line voltage".In this type of device, tripping requires anauxiliary source of energy that is independent ofthe fault current. The source is generally theprotected circuit. When the circuit is energised,the RCD is supplied. If there is no voltage, theRCD cannot operate, but there is no danger.

Type In I∆∆∆∆∆n Standard values of break time (s) and non-actuating time (s)at a residual current (I∆∆∆∆∆) equal to:

5 A, 10 AA A I∆∆∆∆∆n 2 I∆∆∆∆∆n 5 I∆∆∆∆∆n 20 A, 50 A

100 A, 200 A500 A

General Any value Any value 0.3 0.15 0.04 0.04 Maximumbreak times

S u 25 > 0.030 0.5 0.2 0.15 0.15 Maximumbreak times

0.13 0.06 0.05 0.04 Minimum non-actuating times


These devices are designed to operate in spiteof voltage drops as long as the touch voltagecan exceed 50 V (touch voltage limit). Thiscondition is met if a device continues to operatewhen supplied by only two phases with a voltagedrop to 85 V between phases. This is the casefor Vigi modules, the RCDs used with MerlinGerin Compact circuit breakers.

Another distinction for RCDs is whether or nottheir operation is fail-safe. Two types of devicesare considered fail-safe:c those where tripping depends only on the faultcurrent, i.e. all devices without an auxiliarysource are fail-safe,c those, more rarely used, that automatically tripwhen conditions can no longer guaranteetripping in the presence of a fault current (e.g.during a voltage drop to 25 V).

Remarks:

IEC 60364-531-2-2-2 indicates that for devices withauxiliary sources that are not fail-safe, "Their useis permitted if they are installed in installationsoperated by experienced and qualified people".

Standard NF C 15-100, 531.2.2.2 stipulates thatthey may not be used in household installationsor similar applications.

RCDs without auxiliary sources, for which operationdoes not depend on the supply conditions of theprotected circuit, offer high performance and areparticularly well suited to high-sensitivityapplications in residential installations or for finalcircuits that must be reset by unqualifiedpersons, for the reasons listed below:c Final distribution circuits are operated andoccasionally installed by unqualified persons

(without knowledge concerning the installation orawareness of the risks involved).c Final circuits are generally single-phase (Ph/N)circuits and occasionally two-phase (Ph/Ph).c This technique continues to provide protection,even if the neutral or a phase are disconnectedupstream of the RCD.c The devices operate even if the voltage dropsto 0 V.c For additional protection against directcontacts, a high-sensitivity RCD is recognised as

Fig. 12 : Examples of RCDs "without auxiliary source" and "with auxiliary source".

Fig. 11 : The fault current, via the toroid, suppliesenergy to an electromagnet whose moving part is heldby a permanent magnet. When the operating thresholdis reached, the electromagnet counterbalances theattraction of the permanent magnet and the movingpart, drawn by a spring, opens the magnetic circuit andmechanically actuates circuit-breaker opening.

IrIa

A


an efficient means if the PE fails (does not exist,is not connected or breaks). This techniqueoffers a further advantage if the earth resistancerises significantly above 500 ohms (oldinstallations, dry periods, corrosion on the earthelectrode, etc.) in that certain RCDs withauxiliary sources, connected between a phaseand the PE, do not operate correctly under theabove conditions.c This technique is particularly robust given thatno electronic components are continuouslyconnected to the distribution system. The resultis excellent insensitivity to overvoltages andcomponent ageing. (Electronic components, ifpresent, are connected to the secondary of thezero-sequence current sensor and therefore playa role only if a fault occurs and under very lowvoltage conditions.)c This robustness is well suited to installationsthat are not monitored, generally the case forresidential applications.

Operating testAn RCD is a safety device. Whatever thetechnology used, it must always be equippedwith a test system. Although RCDs withoutauxiliary sources are the most reliable,implementation of fail-safe systems on RCDswith auxiliary sources offers an enhanceddegree of safety that does not, however, replacethe periodical test.c Why test RCDs periodically?In practice, a perfectly fail-safe system,particularly concerning internal faults, does notexist. For this reason, in France, RCDs usingauxiliary sources are reserved for industrial andlarge commercial installations and RCDs withoutauxiliary sources for domestic and similarinstallations, which is consistent with theirinherent possibilities described above. In allcases, periodical testing is recommended todetect internal faults.

c PrincipleFor a test, a current is generated that flows in onlyone of the live conductors surrounded by thetoroid, as shown in figure 13. The resistor issized to let through enough current to trip theRCD, taking into account any leakage currentslikely to reduce the test current. The maximumpermissible value is 2.5 times I∆n (for an adjustabledevice, I∆n is the lowest possible setting).The above principle is very common because itis the means to check the entire system, i.e.toroid, relay and breaking device. It is used onearth-leakage protection socket-outlets and onresidual-current circuit breakers with and withoutintegral overcurrent protection. With respect toresidual-current relays with separate toroids, thesame principle is sometimes used. Certain relays,for example Merlin Gerin Vigirex relays, areequipped with a built-in “test” function and alsocontinuously monitor the continuity of the detectioncircuit (toroid/relay link and toroid winding).

Test ItestR

Fig. 13 : Simplified diagram of the circuit for periodicaltests.

4.5 Constraints related to the current sensor

The sensor is a toroidal transformer. It surroundsall the live conductors and is therefore excited bythe magnetic field corresponding to the algebraicsum of the currents flowing in the phases andneutral. The current induced in the toroid and theelectrical signal at the terminals of the secondarywinding are therefore proportional to the residualcurrent.

This type of sensor can detect residual currentsfrom a few milliamperes up to several tens ofamperes.

Cable with a PEThe basic operating principle of an RCD requiresthat the sensor surround only the live conductors.The PE protective conductor must therefore beseparated from the other conductors, as shownin figure 14 next page.

Large conductorsLarge rectangular sensors are available tomeasure the residual current of large conductors(see Fig. 15 next page). Currents should not besummed by using more than one toroid.


If this difficulty is encountered in a main low-voltage switchboard downstream of thetransformer, a toroid may be installed at thehead of the installation, on the earthingconductor of the transformer LV neutral point(see Fig. 16 ). According to Kirchhoff's currentlaw, the residual current detected at (N) isidentical to that at (G) for a fault occurring in theLV system.

Fig. 17 : Incorrect centering in the toroid causes localmagnetic saturation at point A that can result innuisance tripping.

toroid

toroid

Fig. 18 : The toroid must be installed far enough frombends in cables to avoid nuisance tripping.

A3

1

2

Fig. 14 : Running cables with a PE conductor.

Fig. 15 : Rectangular sensors for large cables or bars

HV / LV

RCD

1

N

G

23

RCD

PE

Fig. 16 : Toroid N supplies the same information astoroid G.

High-amperage conductors

To obtain a reliable and linear response from thetoroid, the live conductors must be placed asclose as possible to its centre to ensure that theirmagnetic effects compensate for each otherperfectly in the absence of residual current. Thisis because the magnetic field of a conductordecreases proportionally with the distance. Infigure 17, phase 3 causes local magneticsaturation at point A, i.e. its effect is notproportional. The result is the same if the toroidis positioned near a bend in the cables (seeFig. 18 ). For high currents, parasitic residualinduction may result in a signal on the toroidsecondary and nuisance tripping. The riskincreases when the RCD setting is low withrespect to the phase currents, particularly whena short-circuit occurs.

In difficult cases, (e.g. where Iphase max. / I∆nis high), there are two solutions to avoidnuisance tripping:

c use a toroid much larger than necessary, e.g.twice the size required for the conductors,


c fit a metal sleeve inside the toroid. The sleevemust be made of a magnetic material (soft steel,magnetic sheet metal) (see Fig. 19 ).

When all these precautions are taken, i.e.:c centering of the conductors,c toroid oversizing,c magnetic sleeve,

the ratio Iphase max./I∆n can be as high as 50,000.An RCD with a built-in toroid represents a ready-to-use product for contractors and electricians.The manufacturer carefully designs the completesolution and therefore:c perfectly centers the live conductors and, forlow currents, can design and properly distributea number of primary turns around the toroid,c can "operate" the toroid at higher induction tomaximise the energy measured and therebyreduce sensitivity to stray induction caused byhigh currents.

Magneticsleeve

Fig. 19 : A magnetic sleeve positioned around theconductors inside the toroid reduces the risk of trippingdue to the magnetic effects of high transient currents.

DiscriminationThe goal of discrimination and protectioncoordination is to ensure that only the faulty partof a circuit is de-energised by tripping of theprotective device.

c "Vertical" discriminationThis type of discrimination concerns theoperation of two protective devices installed inseries on a circuit (see Fig. 20 ). Given thetolerances around the RCD thresholds andbreak times, both current and time discriminationare used.

v Current discrimination because, according tostandards, an RCD must operate for a faultcurrent between I∆n /2 and I∆n. In fact, a factorof 3 is required between the settings of twoRCDs to avoid simultaneous operation of the twodevices, i.e. I∆n (upstream) > 3 I∆n (downstream).

v Time discrimination for cases where the faultcurrent suddenly exceeds both rated operatingcurrents (see Fig. 21 next page). It is necessaryto take into account the response time, evenminimal, of all mechanisms, to which it may benecessary to add deliberate time delays.

The double condition to ensure non-tripping ofDa for a fault downstream of Db is:

I∆n (Da) > 3 I∆n (Db) and tr (Da) > tr (Db) + tc (Db)

or tr (Da) > tf (Db)

where:- tr = non-actuating time- tc = disconnection time between the instant theoperating order is given by the measurement

relay to the instant of disconnection (includingthe arcing time),- tf = break time, from detection of the faultthrough to complete interruption of the faultcurrent; tf = tr + tc.

The threshold detection circuits of electronicrelays may exhibit a fault memorisationphenomenon. It is therefore necessary to takeinto account a "memory time", that can bethought of as a virtual increase in the time that acurrent flows, to ensure that they do not operateafter opening of the downstream device.

4.6 Special applications

RCD

RCD

Da

Db

Fig. 20 : Vertical discrimination.


Note:

Particular attention must be paid when determiningdiscrimination conditions for circuit-breakers withadd-on RCDs and residual-current relays usedtogether (see Fig. 22 ). This is because:

- a circuit breaker with an add-on RCD is definedin terms of the non-actuating time (tr),

- a residual-current relay is defined in terms ofthe time between the instant the fault occurs andtransmission of the opening order, to which it isnecessary to add the response time of thebreaking device.

Ifault

a) tr tc

tr tcb)

Fault detected Fault detected

Discrimination

Db open Db open

Ifault

a) tr tc

tr tcb)

No discrimination

Fig. 21 : The time delay of an upstream RCD (a) must take into account the non-actuating time tr and thedisconnection time tc of the downstream RCD (b).

Vigirex RH

tr = 15 ms

tc = 30 ms

tf = 45 ms

Vigicompact

tr = 60 ms

RCD

RCD

Vigirex

tr = 60 ms

tf < 140 ms

tr = 200 ms

Vigicompact

RCD

RCD

Fig. 22 : Two examples of time discrimination between a Vigicompact circuit breaker with add-on RCD and aVigirex relay (Merlin Gerin). Note that these times are much shorter than the authorised actuating times in figure 9.

It is therefore necessary to calculate the successivetf and tr times (at 2 I∆n, the conventional currentfor the non-operating test of delayed RCDs) foreach RCD, from downstream to upstream.

c Horizontal discrimination

Sometimes referred to as circuit selection,stipulated in standard NFC15-100, section535.4.2, it means that an RCD is not necessaryin a switchboard at the head of the installationwhen all the outgoing circuits are protected byRCDs.Only the faulty circuit is de-energised.


The RCDs placed on the other circuits (parallelto the faulty one) do not detect the fault current(see Fig. 23 ).

The RCDs may therefore have the same trsetting.

In practice, horizontal discrimination may presenta problem. Nuisance tripping has beenobserved, particularly on IT systems and withvery long cables (stray capacitance in cables) orcapacitive filters (computers, electronic systems,etc.). Tripping may occur on non-faulty circuits,as shown in figure 24.

Surge arresters

Depending on local utility regulations, RCDs areconnected upstream or downstream of surgearresters. If the RCD is placed upstream, itdetects the current surge produced by lightningand may trip. A delayed or reinforced-immunityRCD is recommended. If the RCD isdownstream, a standard RCD may be used.

Disturbances caused by leakage currents

There are a number of types of leakage currentslikely to disturb RCD operation:c leakage currents at power frequency,c transient leakage currents,c high-frequency leakage currents.

These currents may be natural, flowing throughthe capacitance distributed throughout thecables in the installation, or intentional, i.e. thecurrent flowing through components usedintentionally, namely capacitive filters installedon the supply circuits of electronic devices(computers, variable-speed drives, etc.). Thepurpose of these filters is to bring the devicesinto compliance with the emission and immunitystandards made mandatory by European EMCdirectives.

Fig. 23 : Example of horizontal discrimination.

RCDRCD

c Leakage currents at power frequency(50 or 60 Hz) (see Fig. 25 next page)These currents are generated by the supply sourceand flow through natural or intentional capacitance.

For a single-phase device in a 50 Hz system,continuous leakage currents of approximately0.5 to 1.5 mA per device are measured. Theseleakage currents add up if the devices areconnected to the same phase. If these devicesare connected to all three phases, the currentscancel out when they are balanced (thealgebraic sum is equal to zero).Because of these leakage currents, the number ofdevices that can be connected downstream of anRCD is limited. See Appendix 3 for a comparisonof leakage currents in the different SEAs (TT/TNor IT), which explains why the number of devicesthat may be connected in an IT system is lowerthan in the TT or TN systems.Given that RCD tripping may take place startingat 0.5 I∆n, it is advised, in order to avoidnuisance tripping, to limit the continuous leakagecurrent to 0.3 I∆n for TT and TN systems and to0.17 I∆n for an IT system.

Db

Da

RCD

(A)

(B)

RCD

1

2

3

long cables

Cp

Fig. 24 : In the event of a fault, Da may open instead of Db.


Use of an RCD with a narrow operating range(0.7 I∆n to I∆n) reduces this constraint.A narrow operating range is available from "si"(super-immunised) or Vigirex RCDs from MerlinGerin.

c Transient leakage currentsThese currents appear when energising a circuitwith a capacitive unbalance or during acommon-mode overvoltage (see Fig. 26 ).For example, measurements carried out whenstarting a workstation equipped with a capacitivefilter revealed a transient leakage current withfollowing characteristics:v amplitude of the first peak: 40 Av oscillation frequency: 11.5 kHzv damping time (66 %): 5 periodsRCDs with a certain non-actuating time avoidnuisance tripping caused by this type ofwaveform. Examples are "si" type RCDs(I∆n = 30 mA and 300 mA), Vigirex and S-typeRCDs (I∆n ≥ 300 mA).

c High-frequency leakage currentsHigh-frequency leakage currents (a few kHz upto a few MHz) are caused by the choppingtechnique used by variable-speed drives or theelectronic ballasts of fluorescent lighting. Certainconductors are subjected to high voltagegradients (approx. 1 kV/µs), which generatemajor current spikes through the straycapacitance of circuits.Leakage currents of a few tens or hundreds ofmA can flow (common mode) and be detectedby the RCD, as shown in figure 27 for avariable-speed drive.Unlike the 50 Hz - 60 Hz leakage currents forwhich the algebraic sum is zero, these HF

Flow off ofcurrent generatedby lightning

Surge arrester

A

RCD

B

RCD

Fig. 25 : Depending on local regulations, in aninstallation containing a surge arrester, the RCD maybe placed at A (S-type or immunised RCD) or at B(standard RCD)

Fig. 26 : Leakage current caused by the capacitancedistributed throughout the cables or flowing through theinput capacitors of devices (dotted lines).

Fig. 27 : RCD disturbance caused by high-frequencyleakage currents.

N

Ma

a

N

Device with capacitive filter

currents are not synchronous over all threephases and their sum constitutes a non-negligible leakage current.

In order to prevent nuisance tripping, RCDs mustbe protected against these HF currents(equipped with low-pass filters). This is the casefor industrial RCDs of the Vigirex range and forthe Merlin Gerin “S”, “A si” and “B” type RCDs.

Variable-speed drives

For combinations of RCDs and variable-speeddrives using frequency conversion, it isnecessary to simultaneously take into account anumber of constraints:

c leakage currents when energising,

c continuous leakage currents at 50/60 Hz,

c continuous HF leakage currents,

c special current waveforms for faults at thedrive output,

c current with a DC component for faults on theDC bus.


An analysis of these phenomena and solutionsto satisfy the constraints are presented in detailin Cahier Technique publication no. 204, LVprotection devices and variable-speed drives.

See also Appendix 2, Types of converters andfault-current waveforms.

Uninterruptible Power Supplies (UPS)

In installation with backup sources such asUPSs, the protection system must take intoaccount the different possible configurations. Inparticular operation on AC power or on thebatteries, bypass switches closed or not, etc.

In the example in figure 29, the installation (TTsystem) includes a UPS. If AC power fails, it isnecessary to earth the neutral downstream ofthe UPS (i.e. close contactor K) to ensurecorrect operation of the RCDs.However, this earthing operation is notindispensable to protect persons because:c the installation becomes an IT system and thefirst fault is not dangerous,c the probability of a second insulation faultoccurring during the limited time of operation onbattery power is very low.

Fig. 28 : RCDs with HF-current filtering (VigirexRH99M and RH99P from Merlin Gerin.

Modular version

Switchboard version

Fig. 29 : When loss of AC power is detected, contactor K closes to recreate the TT system downstream of the UPS.

3L

3L

3LN

Automatic bypass switch

3L

3L

N

3L

N

3L

N

AC bypass circuit

N

Priorityloads

Non-priorityloads

NRelay to detect voltage failure

Manual bypass switch

(Maintenance) K

Fault supplied by AC power

Fault supplied by UPS battery


5 Conclusion

At a time when electricity has come to play anincreasingly dominant role in residential,commercial and industrial applications, it isuseful to review and quantify electrical hazardsand provide information on residual-currentdevices (RCD).As for all devices, they have their strong andweak points. Not yet fully perfected, theynonetheless play an increasingly important rolein the protection of life and property. Allindustrialised countries make extensive use ofRCDs, with a variety of system earthingarrangements, in both industry and housing.The following are the most important points to beretained from installation standards and currentpractice.

c For the protection of persons against directcontacts, an RCD is not only very useful, butoften an additional measure required bystandards, whatever the SEA. It is the ultimateline of defence in the protection of human life.

c For the protection of persons against indirectcontacts, an RCD is:v compulsory for the TT system,v necessary for the IT system if there are severalearth electrodes,v recommended for very long circuits on TN andIT systems.

c RCDs also provide protection against:v fires of electrical origin. They are the onlyeffective means to limit fire hazards caused bytracking currents, whatever the SEA,v destruction of machines in the TN system.

Modern RCDs continue to progress in terms ofreliability and immunity to interferencephenomena that are not insulation faults.The purpose of this document is to furtherknowledge of RCDs and thereby contribute tothe safety of life and property.


Appendix 1: calculation of touch voltages

This section indicates briefly how touch voltagesdue to insulation faults are calculated, dependingon the SEA.

UdRd

N

A U0

BC

D

PE

Id

TN system

Fig. 30 : Touch voltage for insulation faults on a TN system.

Ud =0.8 U

2 if R R and Rd = 0o

PE ph= IdU

R Rd R0.8 U

R +Ro

AB CD

o

ph PE=

+ +⇒

In a 230/400 V system, the touch voltage Ud is therefore 92 V. This voltage is greater than theconventional touch voltage limit UL and represents a danger, i.e. the circuit must open.

In general, given the level of the fault current Id, opening can be initiated by the overcurrent-detection devices.

When the resistance values Rph and RPE are high or unknown, RCD protection is required.

For more information, see Cahier Techniquepublication no. 172, Earthing systems in LV.


TT system

Ud

N

PE

Rb Ra

IdU0

Fig. 31 : Touch voltage for insulation faults on a TT system.

N

If

If If

Rb

If

Insulationmonitoringdevice(IMD)

Surgelimiter

U0 321N

PE

Cf

Icn Ic1 Ic2

Cf Cf Cf

Ud ≈ Rb If

Fig. 32 : Touch voltage for insulation faults on an IT system.

IT system

In a 230/400 V system, the touch voltage is approximately 115 V (if Ra=Rb). This voltage isgreater than the conventional touch voltage limit UL and represents a danger, i.e. the circuitmust open.If the earth resistance is approximately 10 Ω, the fault current is approximately 11 A.In general, opening cannot be initiated by the overcurrent-detection devices.Use of an RCD is therefore mandatory.

IdU

R Ro

a≈

+ bUd

RR R

a

a b=

+Uo

Even with high leakage capacitances of approximately 1 µF, the leakage current If for the firstfault is less than 0.1 A.The result is a harmless touch voltage of approximately one volt. Disconnection is notnecessary for the first fault.If a second fault occurs, the situation is that of the TN system.


Appendix 2: types of converters and fault-current waveforms

Fig. 33 : Fault currents corresponding to different semiconductor assemblies (continuation p. 26).

Standard EN50178 (Electronic equipment foruse in power installations) indicates the types ofRCD to use in combination with different

semiconductor assemblies. It also indicates thecorresponding fault-current waveforms.

L

1

2

3

4

5

ILIL

IdN tPE

LIL

IdNPE

L

IL

IdN

PE

L2L3

L1

IL

IdNPE

L

IL

IdN

PE

Id

t

Id

t

IL

t

Id

t

IL

t

Id

t

IL

t

Id

t

IL

t

Single-way connectionwith back - e.m.f. ioad

Three-phase star connection

Two-pulse bridge connection

Two-pulse bridge connectionhalf-controlled

Single-way connection

Connection Fault currentMains current in normal operation


Circuits no. 8 and 9 must be protected by typeAC, A or B RCDs.

Circuits no. 1, 4 and 5 must be protected by typeA or B RCDs.

Circuits no. 2, 3, 6 and 7 must be protected bytype B RCDs.

Examples of loads requiring type A or BRCDs:

c Equipment with single-phase diode rectifiers(circuit no. 4)v Examples include pumps, fans, air-conditioners, lifting and handling equipment, lifts,packing machines, special machines (textile,machine tools, etc.).Power ratings are 0.37 to 2.2 kW for230 V/50 Hz (for higher ratings, the supply isgenerally three-phase).

An insulation fault is possible if a braking resistoris connected to the DC circuit (DC bus). Aninternal insulation fault is highly unlikely.v Power supplies for DC circuitsExamples include welding equipment, batterychargers, electronic devices (PLCs, regulators,telephone exchanges, etc.), excitation windingsof DC motors, electromagnet coils.The maximum power rating is 3 kW (for higherratings, the supply is generally three-phase).

Remark. Most of the time, these devices have anisolating transformer upstream of the rectifier. Inthis case, an insulation fault between the DCcircuit and earth does not cause a fault current. Itis thus possible, for example, to operate with onebattery pole earthed.v Switch-mode power suppliesExamples include computer hardware, stereo andvideo equipment, etc.

7

8

9

6

N

L1

IL

IdL2

PE

Id

t

IL

t

Two-pulse bridge connectionterminal connection between phases

Six-pulse bridge connection (Three-phase bridge connection)

L

A.C. power controllerphase control

A.C. power controllermulti-cycle control

IL

IdNPE

N

L1

IL

IdL2

PE

L3

IL

t

Id

t

IL

IL

t

Id

t

Id

t

LIL

IdNPE

Connection Fault currentMains current in normal operation

Fig. 33 (continuation of page 25) : Fault currents corresponding to different semiconductor assemblies.


c Equipment with single-phase SCR rectifiers(circuit no. 5)v Variable-speed drives for DC motorsThis technique has been largely replaced byfrequency converters, but still exists.Power ratings are less than 10 kW.v Battery chargersThis type of rectifier is used for certain batterychargers, however an isolating transformer isgenerally installed upstream of the rectifier.Consequently, there is no residual current if afault occurs downstream of the rectifier.

Other types of equipment with non-sinusoidal fault currents

c Frequency converters with a single-phasepower supplyThe input circuit is a circuit no. 4. For a fault onthe DC circuit, a type A RCD is suitable.The current waveform for a fault on the output ofa single-phase frequency converter is shown infigure 34. This waveform is not described bypresent standards. The IEC has begun studiesto cover these special cases as well. Even if thiscurrent does not correspond to the waveformindicated for type A RCDs, the type A RCDsfrom Merlin Gerin provide protection.c Frequency converters with a three-phasesupplyThe input circuit is a circuit no. 7 and requires atype B RCD.The current waveform for a fault on the output ofa 3-phase frequency converter is shown infigure 35.A type B RCD is perfectly suited. If there is norisk of a fault on the DC bus, a type A RCD isalso suitable, even if this type of fault currentdoes not correspond to the waveform indicatedfor type A RCDs.

Fig. 35 : Fault current at the output of a three-phasefrequency converter.

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.020.01 0.030 0.04

A

t

(s)

Fig. 34 : Fault current at the output of a single-phasefrequency converter.

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.010

A

t

(s)

0.02


N C

CPE

i1V1

a

N C

CPE

i1

i2

V1

V2

a

a

Fig. 36 : Device connected to a TN system.

Fig. 37 : Device connected to an IT system.

Appendix 3: leakage currents for different systemearthing arrangements

Difference between zero-sequence currentsin TT/TN and IT systemsConsider the simplified diagram of a devicesupplied by a phase and neutral, in the TNsystem. The capacitors C are connectedbetween the live conductors and earth to makethe device immune to power systemdisturbances.

The current measured by the RCD is equal to:i1 = V1 C ωFor an IT system, if the first fault is assumed tooccur on phase 2, the simplified diagram is thatshown below.

The current measured by the RCD is equal to:iT = i1 - i2

where i1 = (V1 - V2) C ω

i2 = V2 C ω

What is more, it is clear that:

V1 = V sinωt

V2 = V sin(ωt - 2π3

)

Calculation of iT:

iT = i1 - i2 = (V1 - 2V2) C ω

i V C sin t - 2sin t -23T = ⎛

⎝⎜

⎞⎠⎟

⎡

⎣⎢

⎤

⎦⎥ω ω ω

π

i V C sin t - 2 sin t cos23

- sin23

cos tT = ⎛⎝⎜

⎞⎠⎟

⎡

⎣⎢

⎤

⎦⎥ω ω ω

π πω

i V C 2 sin t +3

2 cos tT =

⎛

⎝⎜

⎞

⎠⎟ω ω ω

This expression can be written as:

iT = V C ω 2 a (cosα sinωt + sinα cosωt)

iT = V C ω 2 a sin(ωt + α)

where identication gives:

a cosα = 1

a sinα =3

2

hence: a2 cos sin 134

2 2α α+( ) = +

As a result: a =72

i V C 7sin( t )T = +ω ω α

The absolute value of the leakage current is

7 ≈ 2,6 times higher for the first fault on an ITsystem than on a TN system.


Max nb of ballasts TT TN-S ITper si RCD300 mA 300 220 10030 mA 30 22 10

Max nb of loads TT TN-S ITper 30 mA "si" RCDOffice PC * 6 4 2Workstation ** 3 2 1

There is therefore a risk of nuisance tripping forthe first fault in the IT system, i.e. it is necessaryto reduce the number of devices monitored byeach RCD, compared to the number possible inthe TN system. (See the table below).

Limitation of the number of devicesmonitored by each RCD:

Load made up of computers.

*: Includes the PC, a monitor and a laser printer

**: Includes the computer with extensions, a largemonitor and a laser printer.

For smaller configurations, the number of loadscan be increased.

Load made up of lamps with electronic ballasts:


Appendix 4: RCD thresholds and power system voltages

In the United States, certain circuits supplyingsocket-outlets and not equipped with a PEconductor are protected by a GFCI (ground-faultcircuit interrupter) which is a residual-currentdevice. This is required by article 210-8 of NEC,680-10, 511-10. If residual-current protection isprovided, it is built into the socket-outlets and thesensitivity used is 5 mA.

The decision to use a sensitivity of 5 mA(± 1 mA) is not discussed anywhere in detail,however a number of factors explain thedecision.

Note that 120 V distribution in the TN-S systemsignificantly reduces the risks. If a solidinsulation fault occurs in a device and if theresistance of the phase conductors (size, length)is equivalent to that of the return conductors (PEor metal conduit), the touch voltage of theexposed conductive parts (ECPs) on the faultydevice is equal to approximately half the phasevoltage, i.e. 60 V.

This 60 V voltage is close to the 50 V voltagerecognised as being not dangerous(conventional touch voltage limit). Consequently,the standardisation organisation in the USAconsiders that given the characteristics ofLV distribution in North America, additionalprotection against direct contacts is not asnecessary as in three-phase 230/400 V systemswhere the touch voltage of the ECPs of a faultydevice is twice as high.

This explains why, in the USA, protection againstdirect contacts is not mandatory onswitchboards, but only on the socket-outlets ofcertain circuits.

For direct contact with a conductor, e.g. adamaged extension cord, the touch voltage inthe USA is 120 V. The impedance of the humanbody at 120 V is higher than at 230 V andamounts to approximately 2200 Ω (medianvalue). The current that would flow in the bodywould therefore be 120 V/2200 Ω = 54.5 mA.

A 30 mA RCD would operate in 300 ms for acurrent of 54.5 mA (< 2 I∆n) according to theactuating-time tables in the IEC standards. Theperson in question would be severely affected bya current flowing through the body for thisrelatively long time.

In a system with a 120 V phase-to-neutralvoltage, a 5 mA RCD is therefore better becausethe tripping time for the same 54.5 mA current(> 5 I∆n) is only 40 ms. In this case, tripping isas fast as a 30 mA RCD on a 230 V system.

The 5 mA sensitivity used in the USA forprotection against direct contacts in socket-outlets would therefore appear suitable for thetwo-phase TN-S systems (240 V phase-to-phase) used in the USA.

For a three-phase system (230 V phase-to-neutral), the 30 mA sensitivity is better suited toprovide protection against direct contacts, withdevices installed in switchboards or, whereapplicable, in socket-outlets.


Appendix 5: Bibliography

Reference documents:

c IEC 60364, Electrical installations in buildings

c IEC 60479-1, Effects of current on humanbeings and livestock - Part 1. General aspects

c IEC 60479-2, Effects of current on humanbeings and livestock - Part 2. Special aspects

c IEC 60755, General requirements for residualcurrent operated protective devices

c IEC 60947-2, Low-voltage switchgear andcontrolgear, Part 2. Circuit breakers

c IEC 61008, Residual current operated circuit-breakers without integral overcurrent protectionfor household and similar uses (RCCBs)

c IEC 61009, Residual current operated circuit-breakers with integral overcurrent protection forhousehold and similar uses (RCBOs)

c IEC 61200, Electrical installation guide, Part413. Protection against indirect contact -Automatic disconnection of supply

c EN 50178, Electronic equipment for use inpower installations

Schneider Electric "Cahiers Technique"publications

c Uninterruptible static power supplies and theprotection of personsJ.-N. FIORINA,"Cahier Technique" no. 129c Development of LV circuit breakers to standardIEC 60947-2E. BLANC,"Cahier Technique" no. 150c Earthing systems in LVLACROIX and R. CALVAS,"Cahier Technique" no. 172

c Earthing systems worldwide and evolutionsB. LACROIX and R. CALVAS,"Cahier Technique" no. 173c Disturbances in electronic systems andearthing systemsR. CALVAS,"Cahier Technique" no. 177c The IT earth system in LVF. JULLIEN and I. HERITIER,"Cahier Technique" no.178c Cohabitation of high and low currentsR. CALVAS and J. DELABALLE,"Cahier Technique" no. 187c LV protection devices and variable speeddrives (frequency converters)J. SCHONEK and Y. NEBON,"Cahier Technique" no. 204

Other publications

c Electrical installation guideSchneider Electric CITEF

Schneider Electric Service prescriptionF-38050 Grenoble cedex 9Fax: 33 (0)4 76 57 65 28

E-mail : [email protected]

Editor: Schneider ElectricDTP: Axess

© 2

006

Sch

neid

erE

lect

r ic

02-06xxxxx

Documents

Cahier technique no. 114 - pangonilo.com