GB2268011A - Residual current device - Google Patents

Abstract

A residual current device disconnects SCR1, SOL, S1 a mains supply in response to an earth fault current sensed by winding T1 exceeding a pre-determined reference level. An output 81 is indicative of the level of earth fault current and may be connected to a meter or provide a visual warning at a reference level lower than that at which the device trips. The device draws power from the live L and neutral N lines but may alternatively use the earth E line if there is a loss of neutral and warn with LED1. Impedance Z1 prevents the use of the earth path except when neutral is lost. Two test windings T2, T3 connected in antiphase between neutral and earth, with the mid-point connected to live or to a reference voltage, detect a potential difference between neutral and earth lines, and disconnect the supply. Capacitor C1 charges sufficiently to activate the solenoid on firing of SCR1. On Cf charging to the reference voltage level required for disconnection, the fault current must persist for a further fixed time delay determined by Rt Ct before the supply is disconnected. <IMAGE>

Description

A Residual Current Device This invention relates to a residual current device (RCD).

Electronic RCD's have been in use for over thirty years, and during this time have undergone many improvements. Most of these improvements have been brought about to overcome many of the inherent problems of electronic RCD's, examples of some of the more typical problems being; a) Mains power dependency for powering up of electronic circuitry b) Mains power dependency for activation of tripping means c) Nuisance tripping The primary requirement of an RCD is to provide protection to people against electric shock. However, RCD's often trip due to the flow of earth fault currents in circuits and appliances rather than through a person's body, and earth fault currents can build up on a circuit over time, as insulation deteriorates.The user generally does not realise that an earth fault has developed in a circuit until after the RCD has tripped, which usually proves very inconvenient.

According to the present invention there is provided a residual current device including means for disconnecting a mains supply in response to an earth fault current which exceeds a pre-determined reference level, the device further having means providing an output indicative of the level of earth fault current when it is below the pre-determined level.

The output is preferably a voltage whose level corresponds to that of the earth fault current, and which is either fed to a meter and/or fed to a circuit which provides a warning when the earth fault current exceeds a predetermined level.

The advantage of the invention is that since the RCD has means indicating the level of earth fault current that is present on a circuit at any given time, this can facilitate timely maintenance or repair of the faulty circuit or appliance, and reduce the risk of the RCD tripping due to non life threatening situations.

A further disadvantage of conventional RCD's is that most RCD's have circuitry whose response time is proportional to the magnitude of the earth fault current, being shorter for a high magnitude fault current than for a lower magnitude fault current.

While it is desirable that RCD's should respond faster to a high amplitude fault current because the higher the fault current magnitude, the greater the danger it represents. However, it is also desirable that false triggering due to short duration non-earth fault signals be reduced.

Preferably, therefore, the disconnecting means comprises a first circuit means having a response time which varies according to the level of the earth fault current, and a second circuit means which provides a time delay after the first circuit means responds before the mains supply is disconnected, the period of time delay being independent of the level of earth fault current, and the mains supply not being disconnected if the first circuit means determines, before the end of the time delay period, that the earth fault current has fallen below the reference level.

By "response time" above we mean the time taken for the first circuit means to determine that the earth fault current is above the pre-determined reference level.

An embodiment of the invention will now be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a block functional diagram of an RCD typical of the prior art, Figures 2 to 6 are block diagrams illustrating progressive improvements to the prior art ROD, Figure 7 is a block diagram of the processing circuit in the EPU of the prior art RCD of figure 1, Figure 8 is a block diagram of an improved processing circuit, Figures 9(a) and 9(b) are graphs illustrating the operation of the circuit of figure 8, Figure 10 is a block diagram of a further improved processing circuit, and Figure 11 is a block circuit diagram of the embodiment of RCD according to the invention, incorporating all the improvements shown in the preceding figures.

Figure 1 is a block functional diagram of a typical electronic ROD. The basic component parts are a toroidal current transformer CT, a power supply unit PSU, an electronic processing unit EPU, and a tripping mechanism activator TMA.

The PSU provides power from the mains supply, typically by rectification and attenuation of the mains supply as indicated by diode D1 and resistor R1, to provide a low voltage DC supply to the EPU to enable it to function. Thus, the RCD is effectively connected directly across the live and neutral of the mains supply. The mains circuit being monitored by the RCD has its live and neutral current-carrying conductors L and N respectively passing through the CT. Under normal conditions, when a switch S1 is closed, current will flow to the load R from the supply live to supply neutral, with the result that I(live) = I(neutral).

Under an earth fault condition, a current I(fault) will flow to earth, with the result that I(live) will now be greater than I(neutral). This differential current, I(live) - I(neutral), will act as an excitation current for the CT, which will induce a current Ir into winding T1, which will be detected by the EPU. Current Ir is processed by the EPU, and if it is above a predetermined level, the EPU will provide an output to the TMA, thereby causing activation of the tripping mechanism and resultant opening of S1 to disconnect the mains supply.

A test circuit comprising a resistor Rt, a normally open test switch Test 1 and a winding T2 is provided to enable the RCD to be tested by the user.

When the test switch Test 1 is closed, a current flows through RT and T2. This current acts as an excitation current for the CT, producing a resultant output from the EPU which causes activation of the TMA and disconnection of the mains supply.

a) Ameliorating the RCD Power Supply Problem.

One of the main weaknesses of the electronic RCD is that it requires power from the mains supply in order to provide protection against earth fault conditions.

However, if the supply side neutral conductor is broken or left disconnected, the RCD will not be powered up.

Mains live would still be connected to the load when S1 is closed. However, if an earth fault condition now developed at the load, the RCD would not be powered up and would therefore not be able to respond to such a condition.

Figure 2 shows a circuit arrangement which overcomes this loss of neutral, and hence loss of supply, problem. In this circuit, the RCD is still connected to supply live via D1/Rl, but now connected to neutral via D2 and also connected to earth via D3. Under normal conditions, the supply neutral and earth are at the same potential. If the supply neutral is now disconnected, the RCD will be energised via the supply live/earth connections, thereby ensuring continuity of supply to the RCD and enabling it to continue to provide protection in the event of earth fault currents.

The circuit arrangement of figure 2 is designed to provide for those contingent situations of loss of supply neutral. However, a potential drawback of this circuit arrangement is that when the RCD is connected simultaneously to neutral and earth, current can flow from live to neutral, Isn, and from live to earth, Ise.

As the earth and neutral are normally at the same potential, Isn could equal Ise. However, there is no way of predicting the ratio of supply current that will flow in the neutral and earth circuits, and if for any reason there is more impedance in the neutral circuit then in the earth circuit, for example due to component characteristics or quality of connections and terminations, line impedances, etc, then more of the supply current will flow in the earth circuit than in the neutral supply circuit. RCD supply current flowing in the earth circuit gives rise to a new problem, which is explained by way of figure 3.

In this circuit, a main ROD, (MRCD), is shown supplying four separate circuits, each of which is fitted with its own ROD. This type of circuit arrangement is often used in factories, and large buildings, etc, and could in fact comprise of a much larger number of downstream RCD's than the four shown in this example. RCD's 1, 2, 3 and 4 derive their supply currents via the live supply, and via earth and neutral in each case. The supply current requirement for each RCD can be typically of the order of 5mA. If this current flowed to earth instead of neutral in each case, there would be a total of 20mA flowing to earth for these four RCD's, and a greater aggregate of earth current for a larger number of downstream RCD's. The RCD supply current flowing to earth in each case is derived from the MRCD live supply, Ism.However, this current does not flow back through the MRCD via its neutral, with the result that Inm is less than Ism. This represent an earth fault current as far as the MRCD is concerned, with the result that the MRCD could trip if this aggregate earth current exceeds its trip level, and thereby cause loss of power to all of its downstream circuits. MRCD's with rated operating currents of lOmA are now becoming increasingly common throughout Europe in order to provide higher levels of safety. Tripping of MRCD's in such installations could be misconstrued as nuisance tripping, particularly when the MRCD trips when there are no loads connected to the downstream RCD's.

To overcome this problem, it is necessary to ensure that when an RCD is connected to earth and supply neutral, the supply current will flow substantially in the neutral circuit, and only minimally in the earth circuit. In figure 2, therefore, this is achieved by placing an impedance Z1 in the earth circuit of the ROD, such that the RCD supply current will be forced to favour the neutral circuit rather than the earth circuit. By placing the impedance Z1 in series with the earth circuit, Ise is reduced. By suitable selection of impedance Z1, Ise can be reduced substantially below the value of Isn, thereby reducing the risk of tripping upstream MRCD's due to excessive RCD supply currents flowing to earth.

Impedance Z1 can be a resistor, diode, zener diode, etc, or any combination of electronic components used to limit the RCD supply current flow to earth.

Since Figure 2 now provides a diode in the earth and neutral paths, rectification of the mains supply is no longer solely dependent on D1, which could therefore be omitted. However, one advantage of retaining D1 is that the reverse blocking voltage of the overall circuit will be doubled. For example, if the reverse blocking voltage of diodes D1, D2, and D3 is the same at about 1000 volts each, then the retention of diode D1 will provide a total reverse blocking voltage of 2000 volts against high voltage surges appearing between live and neutral and between live and earth. Thus, retaining D1 improves the overall immunity to high voltage surges, and improves overall RCD reliability.

The use of the earth conductor as an alternative supply connection in the event of loss of supply neutral substantially reduces the risk of the RCD becoming inoperable due to such a condition. However, it is possible that the user may fail to connect the earth to the ROD. It is desirable, therefore, to ensure that the necessary earth connection is made to the ROD.

In Figure 2 this has been achieved by connecting the test circuit Test 1 between live and earth, rather than between live and neutral as in figure 1. It is a requirement of most wiring regulations that RCD's be functionally tested by way of operation of the test button after installation. With this test circuit arrangement, the RCD could not be functionally tested and verified to be operating correctly unless the earth connection was made. To limit or minimise the level of test current flowing to earth during test operations, (and thereby preventing tripping of upstream MRCD's), the test winding could comprise of numerous turns on the current transformer CT used in conjunction with a high ohmic value test resistor Rt, whilst still providing the required ampere turns of the test circuit.

A loss of neutral condition is clearly undesirable, as it presents a potential hazard. It would therefore be advantageous if the user could be alerted to such a condition. An indication of loss of neutral is facilitated by the circuit of figure 2 by connecting a light emitting diode LED1 in series with D3 and Z1. When the supply neutral is connected to the circuit as normal, minimal RCD supply current will flow in this circuit due mainly to the impedance of Z1.

However, when supply neutral is disconnected, the RCD supply current will now flow in the earth circuit, causing LED1 to light up, and thereby indicating the loss of neutral condition. Instead of being a light emitting diode LED1 can be any other indicating means, such as a neon light or filament light, and if a suitable filament light or neon light is used, it can simultaneously perform the roles of Z1 and LED1, obviating the need for a separate Z1.

Because a loss of neutral condition is potentially hazardous, the user may prefer the RCD to automatically disconnect the supply from the load in the event of loss of supply neutral, thereby removing the potential hazard from the load circuit. This type of automatic response is facilitated by the circuit of figure 4.

In figure 4, an additional winding is formed on the current transformer CT, designated T3. One end of T3 is connected to mains live L and the other end is connected to earth E via resistor Re and normally closed switch, Test 3. This part of the arrangement is referred to as the earth circuit.

One end of T2 is connected to mains live L, and the other end is connected to mains neutral N via a resistor Rn and a normally closed switch, Test 2. This part of the arrangement is referred to as the neutral circuit.

Winding T1 is connected to the EPU which is responsive to any current flowing in winding T1.

In winding T2, a current I2 will flow from L to N via Rn. In the absence of any current flowing in T3, I2 will on its own induce a current Ir into winding T1 which will be detected by the EPU.

In winding T3, a current I3 will flow from L to E via Re. In the absence of any current flowing in winding T2, I3 will on its own induce a current Ir into winding T1 which will be detected by the EPU.

If the earth and neutral points are at the same potential, which they should be in accordance with the wiring practices of many European countries, I2 will be of the same magnitude as I3 provided that the total resistance of Rn and T2 equals that of Re and T3, given that both currents are derived from the same live potential.

The windings T2 and T3 are wound on the current transformer in opposite directions, ie in anti-phase relationship. Then, assuming the number of turns in T2 equals the number of turns in T3 and that I2=I3, the vector sum of the currents induced into the secondary winding T1 by the currents T2 and I3 on the primary side of the transformer CT will be zero, with the result that there will be no net current flowing in the secondary winding T1.

However, the number of turns in T2 and T3 need not be equal provided the number of turns in windings T2 and T3 and the ohmic values of Rn and Re are such that the ampere turns in the earth and neutral circuits are equal when earth and neutral are at the same potential, that is to say provided that (I2 x turns in T2) = (I3 x turns in T3). Since the windings T2 and T3 are wound in such directions as to be in anti-phase the vector sum of the earth and neutral circuit ampere turns will be zero when the earth and neutral points are at the same potential. Once again, in order to minimise the current flowing from the RCD to earth, multiple turns of T3 can be used with a high ohmic value Re.

If the earth connection is now removed, I3 will cease to flow, and I2 will be the value of net current flowing in windings T2 and T3, and a resultant current Ir will be induced into winding T1 which will be detected by the EPU, thereby causing disconnection of the mains supply. Thus, this circuit can detect a missing earth condition.

If the neutral connection is now removed, but the earth connection remains connected, I2 will cease to flow, and I3 will be the value of net current flowing in windings T2 and T3, and a resultant current Ir will be induced into winding T1 which will be detected by the EPU, therebv causing disconnection of the mains supply.

Thus, this circuit can detect a missing neutral condition.

Whilst the earth and neutral points will normally be at the same potential, conditions can arise in the mains supply where the neutral rises to some potential greater than earth potential. Where the differential in earth and neutral potentials becomes significant, it can be desirable to detect such conditions. If the earth and neutral are not at the same potential, I2 will not equal I3, with the result that the vector sum of these two currents will be greater than zero. This resultant net current will induce a current Ir into winding T1 which can be detected by the EPU, thereby causing disconnection of the mains supply. Thus this circuit can detect a difference between earth and neutral potentials.

If the supply side live and neutral connections are reverse wired, the central point of windings T2 and T3 will now be at neutral potential, whilst the other end of winding T2 will be at live potential. A current I2 will flow in winding T2. However, one end of winding T3 will now be at neutral potential whilst the other end will still be at earth potential, therefore no current will flow in winding T3. As I2 will be the only current flowing in the circuit, it will be an out of balance current, and will therefore induce a current Ir into winding T1 which will be detected by the EPU, thereby causing disconnection of the mains supply. Thus this circuit can detect a reverse wired live and neutral.

Because the switch Test 2 connected into the neutral circuit is normally closed it has no bearing on the detection of missing earth, missing neutral, high neutral potential with respect to earth, or reverse live - neutral sensing. However, by opening switch Test 2, I2 can be made to cease to flow. I3 will now be the net current flowing in windings T2 and T3, and a resultant current Ir will flow in winding T1 which will be detected by the EPU, thereby causing disconnection of the mains supply. Thus, by provision of a test switch Test 2, the integrity and correct functioning of the earth circuit, windings T3 and T1, and the RCD can be checked.

Because the switch Test 3 connected into the earth circuit is normally closed it has no bearing on the detection of missing earth, missing neutral, high neutral potential with respect to earth or reverse live - neutral sensing. However, by opening switch Test 3, I3 can be made to cease to flow. I2 will now be the net current flowing in windings T2 and T3, and a resultant current Ir will flow in winding T1 which will be detected by the EPU, causing disconnection of the mains supply. Thus, by provision of a test switch Test 3, the integrity and correct functioning of the neutral circuit, windings T2 and T1, and the RCD can be checked.

With this circuit arrangement, the RCD would trip automatically in the event of mains wiring faults, such as, loss of supply neutral or earth, high neutral potential with respect to earth, or reverse live neutral connection.

Automatic tripping in the event of loss of earth connection is particularly important in those applications where an earth connection to equipment or appliances is an essential requirement, as this new RCD would prevent such equipment or appliances being used unless an earth connection was provided to the ROD.

One problem arising from the circuit of figure 4 is that windings T2/T3 are at mains live potential, and winding T1 is at some lesser potential by virtue of the voltage dropped across the PSU, such that a substantial potential difference exists between these two sets of windings. This can give rise to problems of breakover voltages between the two sets of windings if they are overlaid, or wound very close together. If such a voltage breakover occurred, the EPU components or electronic circuitry could be damaged as a result of the high mains voltages. This problem is normally overcome by isolating or insulating the windings to prevent voltage breakover, which gives rise to problems of space, wire size, etc.

In the present case this high voltage problem may be substantially overcome by connecting the junction of windings T2/T3 to the junction of D2 and Z1, as shown by the dashed line in figure 4, instead of directly to mains live as shown by the solid line. The junction of D2 and Z1 represents the common or 0 volts line of the EPU. Winding T1 is also connected to this common line via the EPU, therefore the two sets of windings T2/T3 and T1 are now substantially at the same potential, obviating the need for high voltage insulation or protection between the two sets of windings.

b) Ameliorating the TMA Energy Problem.

Figure 5 shows a typical circuit arrangement for activation of a tripping mechanism to cause disconnection of the mains supply from a load during earth fault conditions. In this circuit, when the earth fault current exceeds a predetermined level, a voltage signal from the EPU turns on silicon controlled rectifier SORT, effectively connecting a solenoid SOL across the mains supply, and S1 opens as a result of the electromagnetic energy developed in the solenoid. Once SCR1 is turned on, the solenoid needs to be capable of withstanding the power developed within it until such time as the contacts have opened.This requires the solenoid to have a reasonably high impedance at full mains voltage, and to be able to withstand the heat generated in it during multiple operations. (Most RCD standards require the RCD to be able to withstand up to one thousand repetitive tripping operations in quick succession, which can result in a substantial build up of heat within the solenoid windings at full mains voltage). The current limiting impedance in the solenoid results in a substantial reduction in solenoid energising current at reduced mains voltage, with the result that the solenoid ceases to be able to generate sufficient energy to operate the tripping mechanism at low mains voltage.

It is desirable that the RCD should be able to operate effectively at mains voltage levels which are below their nominal level but still potentially hazardous. This would require the tripping mechanism to be operable at mains voltage levels from about 100 volts. However, if the solenoid as shown in figure 5 can operate effectively at 100 volts, it will have to dissipate the surplus power generated at full rains voltages of up to 250 volts; some 2.5 times the 100 volts power level, which can give rise to heat dissipation and resultant winding insulation breakdown problems.

A second problem that arises with mains operated solenoids is the occurrence of a very low impedance or direct short circuit from live to earth at the load. If R(fault) had a very low impedance, which can occur for example when live is shorted directly to a metal frame, then the resultant massive fault current flow I(fault) could result in a substantial voltage drop in the mains supply to the ROD. In exceptional circumstances, this could result in the RCD supply voltage being reduced from 240 volts to less than 100 volts. In such circumstances, it is possible that an activated solenoid could not generate sufficient energy to trip the ROD, and a most hazardous situation would be sustained until such time as alternative protection means came into operation, such as fuses or circuit breakers for overcurrent protection. Both of these problems are overcome by the arrangement of figure 6.

In this circuit, capacitor C1 is charged to a voltage level equal to the breakover voltage of zener diode ZD2. When the earth fault current exceeds the predetermined level, the EPU sends a voltage signal to turn on SCR1. Capacitor C1 now discharges through the solenoid, causing disconnection of the tripping mechanism. The solenoid energising current is a function of the solenoid impedance and the voltage developed across C1, which in turn is determined by the clamping voltage of ZD2. With this circuit arrangement, capacitor C1 will always charge up to the same voltage level once the mains supply voltage exceeds the zener breakover voltage.Components ZD2, C1 and R1 are chosen to ensure that C1 can acquire a sufficient charge to energise the solenoid at the required minimum mains operating voltage. Mains voltages above this level will not result in any increased current through the solenoid, as the voltage across C1 will be clamped by ZD2. Thus the design of the solenoid can now be optimised on the basis of a relatively consistent energising current and voltage, which obviates the need for surplus power handling capabilities in the solenoid whilst still achieving the desired objective of low and high supply voltage operation. Zener diode ZD2 can be an internal voltage regulator of the EPU.

In figure 6, there is also an impedance Z2 connected between R1 and the EPU. The reason for having this is best explained by first assuming that it is omitted. In such case, two problems could arise, as follows.

1) A situation could arise where a standing earth fault could exist at the load such that a fault current starts to flow as soon as S1 is closed. This fault will be detected by the EPU, which will send a signal to the SCR. However, it may be possible that due to the speed of response of the EPU, C1 would have had insufficient time to acquire sufficient charge to activate the solenoid. In such circumstances, C1 would be discharged through the solenoid, and would be unable to cause disconnection of the mains supply. The EPU could continue to trigger the SCR, effectively keeping C1 from acquiring sufficient charge to activate the tripping mechanism.

2) It is possible that the EPU maximum operating voltage could be less than the voltage required to operate the solenoid, in which case, the EPU maximum operating voltage would become the effective limit of the voltage to which C1 would have to be clamped. In such a condition, a high voltage solenoid would not be compatible with a low voltage EPU.

Both of the above problems are resolved by the impedance Z2 connected between R1 and the EPU.

Capacitor C1 can charge up to a voltage level determined by the breakover voltage of ZD2. If this voltage is in excess of the EPU maximum operating voltage, the excess voltage can be developed across impedance Z2, thereby protecting the EPU. Z2 can be another zener diode, or a dropper resistor. This circuit arrangement would facilitate an EPU with a relatively low maximum operating voltage (e.g. 5 volts) being operated with a relatively high solenoid operating voltage (e.g. 100 volts), the surplus voltage being dropped across impedance Z2.

Zener diode ZD2 can comprise one or more diodes in order to achieve the required operating voltage levels for the solenoid. Again, the voltage across C1 will be clamped at the optimum solenoid operating voltage once the mains supply exceeds the clamping voltage of ZD2. The EPU may have its own internal voltage regulator, thereby clamping its own maximum voltage via Z2.

Problems 1 and 2 above are now solved because most of the supply current Is will be diverted to capacitor C1, enabling C1 to charge up rapidly to the clamping voltage of ZD2. Once C1 has reached this level, Is will be diverted via impedance Z2 to the EPU, thereby providing power to the EPU. Any additional surplus supply current will flow via ZD2 back to the supply. The effect of Z2 therefore is to prioritise the charging of C1 over the powering up of the EPU, thereby ensuring that when the EPU turns on SCR1, C1 has acquired the necessary charge to ensure activation of the solenoid and tripping mechanism.

c) Nuisance Tripping Problems.

Nuisance tripping results when the RCD trips in response to non earth fault conditions, such as mains supply disturbances, etc. Most such occurrences of nuisance tripping usually result from the SCR being turned on by: 1) high voltage spikes causing the SCR to turn on due to the sudden appearance of a very high voltage across the SCR, or 2) false triggering by the EPU.

Both of these problems need to be minimised in order to reduce the incidents of nuisance tripping.

1) In the circuit configurations of figures 1 to 5, SCR1 was connected directly to mains live, with the result that it was subjected to voltage spikes appearing on the mains supply. The sudden appearance of a high voltage spike across the SCR could cause the SCR to turn on in the absence of a triggering signal from the EPU, thereby causing nuisance tripping. The circuit arrangements of figure 6 substantially overcomes the problem of SCR turn on due to high voltage spikes. In figure 6, SCR1 is isolated fron mains live by D1 and R1, and from mains neutral by D2.It is further isolated from earth by D3 and Z1, and all of these components act as impedances which buffer SCR1 from high voltage spikes appearing on the mains supply.

2) The EPU normally gives an output signal to SCR1 in response to an input from the CT due to an out of balance current caused by an earth fault condition. The response time of the EPU is normally related to the magnitude of the fault current, in that the EPU will take longer to process a fault current level that is slightly above the EPU reference level than it will for a fault current level that is substantially above the EPU reference level. This means that a high amplitude signal from the CT can result in a virtually immediate response from the EPU, even when the CT output signal is of very short duration. A high amplitude short duration signal is unlikely to be due to a steady state earth fault condition, and is more likely to be caused by factors such as mains voltage disturbances, highly reactive loads, high inrush load currents, etc.

Most EPU's have processing circuitry whose response time is proportional to the magnitude of the CT input signal. Figure 7 shows a typical example of such a circuit.

Under no fault current conditions, the capacitor Cf is discharged, the output of the comparator Comp 1 is LOW because of the presence of a positive value Vref on its -ve input terminal, and the SCRl remains turned off.

When a fault current flows, the resultant input signal from the CT is amplified and rectified at 70 and smoothed by capacitor Cf. When the voltage across Cf exceeds the reference level on the comparator Comp 1, the comparator output goes HIGH, turning on SCR1 and thereby activating the solenoid SOL and tripping mechanism. It can readily be seen that the response time of the circuit will be shorter for a high magnitude fault signal than for a lower magnitude fault signal, and that for any given value of Cf, the response time will be dependent on the magnitude of the fault signal.

It is desirable that RCD's should respond faster to a high amplitude fault current because the higher the fault current magnitude, the greater the danger it represents. However, it is also desirable that false triggering due to short duration non earth fault signals be reduced. This could be facilitated by providing an additional, but precisely adjustable, time delay to the ROD. Such an arrangement is shown in figure 8.

In this circuit, under no fault conditions, Cf is discharged, Comp 1 output is HIGH because the positive Vref is now applied to the +ve input terminal, with the result that transistor Trl is turned on, and Ct is held discharged. The output of a second comparator Comp 2 is therefore LOW. When a fault current flows, the resultant input signal from the CT is amplified and rectified at 70, and smoothed by capacitor Cf. When the voltage across Cf exceeds Comp 1 reference voltage, Comp 1 output goes LOW, turning off Trl. Ct now charges up via Rt towards +Vcc. When the voltage across Ct exceeds Comp 2 reference voltage, Comp 2 output goes HIGH, turning on SCR1, thereby causing activation of the solenoid SOL and tripping mechanism.For any given value of Ct and Rt, the rate of charge of Ct through Rt towards +Vcc is constant, because +Vcc is also fixed, and hence the time constant of the circuit will also be fixed and repetitive.

The additional time delay circuit comprising Ct, Rt and Comp 2 provides three distinct advantages over the more conventional delay circuit of figure 7, as follows.

1) The additional time delay is independent of the magnitude of the fault current level.

2) The additional time delay is precisely adjustable by selection of the values of Ct and Rt. Rt may be a fixed or variable resistor, as desired, thereby permitting designer and/or user selection of fixed time delay periods which are likely to exceed the duration of most commonly experienced transient and spurious mains and load conditions.

3) The SCR will not be triggered unless the duration of the fault current exceeds the time delay set by Ct and Rt. If the voltage across Cf falls below Comp 1 reference voltage, Trl will be turned on again and will immediately discharge Ct, preventing Comp 2 output going high and triggering SCAR1, thereby automatically terminating the timing cycle in the event of the fault condition not being sustained for the full duration of the fixed time delay period.

In addition, the requirement for the RCD to have a faster response time to high level fault currents has been maintained by retaining the circuit arrangement of Cf with its associated amplifier and rectifier, thereby ensuring that the commencement of the timing cycle of the second timing circuit is related to the magnitude of the sustained earth fault current which produces a resultant output from the CT. This relationship is demonstrated by way of an example in figures 9A and 9B.

In figure 9a, a fault current of 30mA is flowing. Vref is set to a level equivalent to 20mA. Cf starts to charge up towards the rectified output voltage VF of the amplifier at a rate dependent on the magnitude of VF. When VCF exceeds Vref, in this case after approximately 23mSec, Comp 1 output goes low, turning off TR1 and permitting Ct to charge via Rt towards +Vcc.

After the fixed time delay of Ct/Rt has expired, in this case 12mSec, Comp 2 output goes high, turning on the SCR.

In figure 9b, a fault current of 150mA is flowing. Vref is still set to a level equivalent to 20mA. However, Cf now has a much higher aiming voltage, and charges up at a much faster rate then in the case of the 30mA fault current. VCF now exceeds Vref in a much shorter period, in this case approximately SmSec, and activates the second time delay circuit. 12mSec later, the SCR is triggered.

The overall response time for the 30mA and 150Ma faults are approximately 32mSec and 17mSec respectively, maintaining a fast response to the higher level fault current. However, in each case, the fault current would have had to continue to flow for the overall periods of 32mSec and 17mSec respectively in order to ensure activation of the tripping mechanism, otherwise the fixed timing cycle would have been terminated, thereby minimising the risk of nuisance tripping.

Thus, the requirement for substantial immunity against false triggering of the SCR has been met by the reconfiguration of the EPU as shown in figure 6 and by the provision of an additional time delay which is independent of the CT output signal magnitude.

The DC voltage developed across the filter capacitor Cf is a measure of the earth fault current flowing in the mains circuit. As shown in figure 10, this voltage is advantageously fed to a DC amplifier 80 and made available as a separate DC output voltage 81 from the EPU. The output 81 can be used to indicate to the RCD user the presence of the earth fault current in the mains circuit at any given time, and by measuring the output voltage 81 the level of the earth fault current can be ascertained. The measuring means can be an external meter which monitors and measures the DC voltage output 81.The indicating means can be an LED which lights up when the earth fault current exceeds a predetermined level, as determined by the user but less than the level at which the RCD triggers, thereby acting as an early warning system. Examples of these are shown in figure lo.

As stated above, the voltage across Cf is fed to the DC amplifier 80 whose output voltage 81 is available as a voltage equivalent of the earth fault current. This voltage is also fed to a comparator, Comp 3. Under no fault conditions the voltage output 81 is zero and the Comp 3 output is low. When the voltage output 81 exceeds the voltage at the junction of the resistors Rx and Ry, Comp 3 output goes high, lighting the LED. This could also be a buzzer or any other kind of voltage operated alarm or signalling device. Rx or Ry can be variable resistors, which can be adjusted to set a desired reference level for Comp 3, thereby facilitating user setting of the level above which the LED should light and provide an earth fault current alert.

Alternatively or as well, the DC output amplifier 80 can be scaled to provide an output voltage for an external meter (not shown), which is readily related to the fault current, e.g. lmA fault current = l0mV DC output. Thus a DC output of 1 volt would indicate a fault current of lOOmA.

Figure 11 is a block circuit diagram of the embodiment of RCD according to the invention, incorporating all the improvements shown and described in the preceding figures.

It is to be understood for the purposes of compactness, the amplifier and rectifier 70, the comparators Comp 1 and Comp 2, and the amplifier 80 providing the DC output signal 81 have all been encapsulated into a single customised integrated circuit IC1. The circuit comprising the comparator Comp 3, the resistors Rx and Ry and the LED is not shown in figure 11, but it may be assumed to be present if desired, connected to the output 81. A meter may also be connected to the output 81, as discussed above.

Power is supplied to the load R via switch S1, the load-carrying conductors L and N being passed through the current transformer CT. The RCD electronic circuitry is connected to the mains supply, comprising live, neutral and earth, the mains supply being rectified and attenuated by D1, R1, ZD2, D2, D3 and Z1.

IC1 has an in-built voltage regulator, which clamps the voltage across IC1 to approximately 5 volts. If the solenoid can be operated directly from this 5 volt supply, then Z2 could be replaced with a short circuit, and ZD2 could be omitted. Capacitor C1 charges up to the voltage level at the junction of R1 and Z2.

In the event of an earth fault current flowing at the load, an output current will be produced by the CT, and will be developed across R3. Capacitor C5 acts as a high frequency filter to attenuate high frequency components that may be produced by the CT. The voltage across R3 is amplified and rectified by IC1, and smoothed by filter capacitor Cf. When the voltage across Cf exceeds an internal reference level, a timing cycle determined by the values of Rt and Ct commences, and on completion of this timing cycle a voltage signal is applied to the gate of SCR1. Capacitor C2 provides some noise immunity to SCR1, and reduces the likelihood of false triggering of SCR1, particularly during the power up period of the circuitry. When SCR1 turns on, C1 discharges through the solenoid SOL, causing activation of the tripping mechanism and opening of S1.

Z2 can have an impedance from zero ohms upwards, and is provided to enable C1 to charge up to a voltage sufficiently high to ensure activation of the solenoid and tripping mechanism at a required voltage. C6 provides additional filtering of the 5 volts supply to IC1, if necessary. The mains monitoring portion of the circuit comprises Rn/T2 and Re/T3, as earlier described.

The ampere turns in the earth and neutral circuits will be the same when earth and neutral are at the same potential, and due to the phase relationship of windings T2 and T3, will cancel to give zero output from the CT under normal conditions. Switches Test 1, Test 2 and Test 3 can be used for test purposes, and when either of these is operated, or in the event that either the earth or neutral connection is missing, or the neutral potential exceeds the earth potential by a predetermined level, or that the live and neutral are connected in reverse, a resultant out of balance current will produce an output from the CT which will be detected by the IC and will cause activation of the tripping mechanism. The junction of T2/T3 is shown connected directly to mains live.However, the dashed extension from this junction shows the alternative arrangement, discussed in relation to figure 4, of connecting this junction to the 0 volts common line of the RCD circuit if it is desirable to avoid voltage breakover problems between the two sets of windings T2/T3 and T1. A DC output voltage 81 proportional to the earth fault current flowing in the mains circuit is provided. As mentioned above, this can be fed to an external meter for measuring purposes, or to a comparator for level indicating purposes as shown in figure 10.

The above described RCD has the following advantages: 1. It provides for the RCD to be powered up in the event of loss of neutral, or in the event of poor quality neutral connections or terminations to the ROD.

2. It incorporates an impedance in the earth supply circuit to substantially minimise the supply current flowing to earth when the neutral connection is present, thereby reducing the risk of upstream RCD's tripping as a result of downstream RCD's being connected to the earth terminal.

3. It provides for effective and consistent operation of the solenoid and tripping mechanism over a wide range of supply voltages by stabilising the voltage level at which the solenoid is required to operate.

4. It overcomes the problems of excessive power dissipation in the solenoid when operated at full mains supply voltage, thereby resulting in reduced heat generation during solenoid operation and reduced risk of resultant solenoid windings insulation breakdown.

5. It substantially reduces the problems of nuisance tripping by isolating the solenoid and SCR from the mains supply by way of impedance and attenuating circuitry, thereby reducing the likelihood of the SCR being triggered by high voltage spikes and surges on the mains supply.

6. It substantially reduces the problems of nuisance tripping by the provision of an additional programmable fixed time delay which is independent of the magnitude of the earth fault current.

7. It incorporates an additional fixed time delay circuit which automatically terminates the timing cycle in the event of the fault condition not being sustained for the full duration of the fixed time delay period, thereby facilitating designer and user selection of fixed time delay periods which are likely to exceed the duration of most commonly experienced transient and spurious mains and load conditions.

8. It incorporates circuitry to ensure that an earth connection has been supplied to the ROD, thereby ensuring that such earth connection is available for the earthing of equipment and appliances connected to the ROD.

9. It incorporates a mains monitoring circuit which is responsive to all of the conditions: loss of earth, loss of neutral, reverse live-neutral connection, high neutral potential with respect to earth potential.

10. It has a mains monitoring circuit which produces out of balance currents in a current transformer in response to adverse conditions of the mains supply thereby enabling such fault conditions to be processed in like manner to fault conditions arising from earth fault currents.

11. It comprises a mains monitoring circuit which does not result in automatic disconnection of the mains supply due to transient or short duration adverse conditions unless such adverse mains supply conditions continue for a period in excess of the combined time delay of the electronic sensing circuitry, such adverse mains supply conditions producing an out of balance current in the current transformer with the result that adverse mains conditions are processed in like manner to earth fault conditions by the ROD, thereby ensuring that all of the inherent improvements in the RCD against nuisance tripping apply equally to adverse conditions of the mains supply.

12. It comprises a mains monitoring circuit which does not require active electronic components such as semiconductors to sense such adverse conditions of the mains supply, etc, thereby providing for improved inherent reliability of the mains monitoring circuitry.

13. It provides verification of the connection of an earth conductor to the ROD.

14. It provides for the optional use of three separate test switches to ensure the correct operation and integrity of the RCD and its associated mains monitoring circuitry.

15. It incorporates an output voltage signal level which is proportional to the earth fault current flowing in the mains circuit, and which can be used for earth current measuring or indicating purposes, and which provides a warning of the presence of such earth fault currents thereby facilitating preventative maintenance prior to the RCD tripping and causing undesirable disconnection of the mains supply.

16. It comprises circuitry which prioritises the charge on a capacitor to ensure that such charge has reached a level sufficient to activate a solenoid and tripping mechanism in the event of the RCD being switched on to a standing earth fault condition.

Claims (10)

CLAIMS:

1. A residual current device including means for disconnecting a mains supply in response to an earth fault current which exceeds a pre-determined reference level, the device further having means providing an output indicative of the level of earth fault current when it is below the pre-determined level.

2. A residual current device as claimed in claim 1, further including means responsive to said output for providing a visual warning when the earth fault current is above a further reference level less than the said pre-determined level.

3. A residual current device as claimed in claim 1 or 2, wherein the device derives its power from the mains and is connected on one side to mains live and on the other side via respective circuit paths to both mains neutral and mains earth so that power can be obtained via the earth connection if there is a loss of neutral.

4. A residual current device as claimed in claim 3, wherein the impedence in the path to earth is greater than that in the path to neutral so as to favour current flow in the path to neutral.

5. A residual current device as claimed in claim 4, wherein the path to earth includes a visual indicator which is energised by flow of current in such path if there is a loss of mains neutral.

6. A residual current device as claimed in any preceding claim, wherein the mains neutral and live current-carrying conductors are inductively coupled to a current transformer having a secondary winding such that when the currents in the two conductors are the same the vector sum of the currents induced into the secondary winding is zero, the current transformer further having a first primary winding connected on one side to earth via a first normally-closed test switch and on the other side to a reference voltage, and a second primary winding connected on one side to neutral via a second normally-closed test switch and on the other side to the said reference voltage, the first and second primary windings having the same number of ampere-turns and being connected in antiphase.

7. A residual current device as claimed in claim 6 when directly or indirectly dependnet on claim 3, wherein the said reference voltage is the voltage at the junction of the neutral and earth paths.

8. A residual current device as claimed in any preceding claim, wherein disconnection of the mains supply is effected by energisation of a solenoid, the maximum voltage across the solenoid being determined by a capacitor in parallel with a voltage limiting device.

9. A residual current device as claimed in any preceding claim, wherein the disconnecting means comprises a first circuit means having a response time which varies according to the level of the earth fault current, and a second circuit means which provides a time delay after the first circuit means responds before the mains supply is disconnected, the period of the time delay being independent of the level of the earth fault current, and the mains supply not being disconnected if the first circuit means determines, before the end of the time delay period, that the earth fault current has fallen below the reference level.

10. A residual current device, substantially as described herein with reference to the accompanying drawings.