Editing Residual-current device (section)

==Characteristics==

===Differences in disconnection actions===
Major differences exist regarding the manner in which an RCD unit will act to disconnect the power to a circuit or appliance.

There are four situations in which different types of RCD units are used:

# At the consumer power distribution level, usually in conjunction with an RCBO resettable circuit breaker;
# Built into a wall socket;
# Plugged into a wall socket, which may be part of a power-extension cable; and
# Built into the cord of a portable appliance, such as those intended to be used in outdoor or wet areas.

The first three of those situations relate largely to usage as part of a power-distribution system and are almost always of the ''passive'' or ''latched'' variety, whereas the fourth relates solely to specific appliances and are always of the ''active'' or ''non-latching'' variety.  ''Active'' means prevention of any ''re-activation'' of the power supply after any inadvertent form of power outage, as soon as the mains supply becomes re-established; ''latch'' relates to a ''switch'' inside the unit housing the RCD that remains as set following any form of power outage, but has to be reset manually after the detection of an error condition.

In the fourth situation, it would be deemed to be highly undesirable, and probably very unsafe, for a connected appliance to automatically resume operation after a power disconnection, without having the operator in attendance{{dash}}as such, manual reactivation of the RCD is necessary.

The difference between the modes of operation of the essentially two different types of RCD functionality is that the operation for power distribution purposes requires the internal latch to remain set within the RCD after any form of power disconnection caused by either the user turning the power off, or after any power outage; such arrangements are particularly applicable for connections to refrigerators and freezers.

Situation two is mostly installed just as described above, but some wall socket RCDs are available to fit the fourth situation, often by operating a switch on the fascia panel.

RCDs for the first and third situation are most commonly rated at 30{{nbsp}}mA and 40{{nbsp}}ms. For the fourth situation, there is generally a greater choice of ratings available{{dash}}generally all lower than the other forms, but lower values often result in more nuisance tripping.  Sometimes users apply protection in addition to one of the other forms, when they wish to override those with a lower rating. It may be wise to have a selection of type four RCDs available, because connections made under damp conditions or using lengthy power cables are more prone to trip-out when any of the lower ratings of RCD are used; ratings as low as 10{{nbsp}}mA are available.

===Number of poles and pole terminology===
The number of poles represents the number of conductors that are interrupted when a fault condition occurs. RCDs used on single-phase AC supplies (two current paths), such as domestic power, are usually one- or two-pole designs, also known as [[single-pole|single-]] and [[double-pole]]. A single-pole RCD interrupts only the energized conductor, while a double-pole RCD interrupts both the energized and return conductors. (In a single-pole RCD, the return conductor is usually anticipated to be at [[ground potential]] at all times and therefore safe on its own).

RCDs with three or more poles can be used on three-phase AC supplies (three current paths) or to disconnect the neutral conductor as well, with four-pole RCDs used to interrupt three-phase and neutral supplies. Specially designed RCDs can also be used with both AC and DC power distribution systems.

The following terms are sometimes used to describe the manner in which conductors are connected and disconnected by an RCD:
:* Single-pole or one-pole – the RCD will disconnect the energized wire only.
:* Double-pole or two-pole – the RCD will disconnect both the energized and return wires.
:* 1+N and 1P+N – non-standard terms used in the context of RCBOs, at times used differently by different manufacturers. Typically these terms may signify that the return (neutral) conductor is an isolating pole only, without a protective element (an unprotected but switched neutral), that the RCBO provides a conducting path and connectors for the return (neutral) conductor but this path remains uninterrupted when a fault occurs (sometimes known as "solid neutral"),<ref>[http://www.voltimum.com.au/content/rcbo-2-pole-and-1pn Explanation on voltimum.com.au, by specialist Ian Richardson].</ref> or that both conductors are disconnected for some faults (such as RCD detected leakage) but only one conductor is disconnected for other faults (such as overload).<ref>http://docs-asia.electrocomponents.com/webdocs/01e3/0900766b801e3b4d.pdf (states that there is "2 pole switching of phase [energized] and neutral [return]", but then only identifies the energized conductor as being protected against "overloads and short circuits").</ref>

===Sensitivity===
RCD sensitivity is expressed as the rated residual operating current, noted ''I<sub>Δn</sub>''. Preferred values have been defined by the IEC, thus making it possible to divide RCDs into three groups according to their ''I<sub>Δn</sub>'' value:
* High sensitivity: 5, 10 and 30{{nbsp}}mA (for direct-contact or life injury protection),
* Medium sensitivity: 100, 300, 500 and 1000{{nbsp}}mA (for fire protection),
* Low sensitivity: 3, 10 and 30{{nbsp}}A (typically for protection of machines).

The 5{{nbsp}}mA sensitivity is typical for GFCI outlets.

===Trip time===
Trip times are how quickly the device reacts to a fault. Devices must meet specific trip times when leakage current is sensed, with times ranging from tens to hundreds of milliseconds, as defined in national and international standards. The amount of time may vary depending upon the amount of current sensed, potentially having to react quicker to larger amounts, and to not react at all if the amount is sufficiently below the device rating.

In contrast to ''general'' types, ''S-type'' (aka ''time delayed'') RCDs include a time delay — a minimum amount of time during which the device will not react — and typically higher maximum reaction time limits. This time delay allows for ''discrimination'' (aka ''selectivity'') which is the principle that where two RCDs are installed in series, in the even of a fault it is desirable for the downstream device (which would typically cover a smaller portion of the installation) to trigger rather than the upstream in order to minimise how much of the electrical installation is disconnected. The upstream device should only trigger if the downstream device fails to react or if the fault is located between the two devices.

''Programmable'' devices may be available for industrial use that can be adjusted to a range of sensitivity values and possibly response times.

===Type (types of leakage current detected)===
IEC Standard 60755 (''General requirements for residual current operated protective devices'') defines the following types of RCD depending on the waveforms and frequency of the fault current:<ref>{{cite book |title=BS 7671 - Requirements For Electrical Installations |edition=18th:2018+A2:2022 |publisher=[[Institution_of_Engineering_and_Technology|IET]] |pages=156–157}}</ref>
;'''Type AC''' [[file:Typ AC-Symbol.png|40px|RCD type AC symbol]]: These trip on alternating sinusoidal residual current, suddenly applied or smoothly increasing.
;'''Type A''' [[file:Typ A-Symbol.png|40px|RCD type A symbol]]: These trip on both alternating sinusoidal residual current and residual pulsating direct current, suddenly applied or smoothly increasing.
;'''Type F''' [[file:Typ A-Symbol.png|40px|RCD type F symbol part 1]]&nbsp;[[file:Typ F-Symbol.png|40px|RCD type F symbol part 2]]: These trip in the same conditions as Type A, and additionally:
:* For composite residual currents, whether suddenly applied or slowly rising, intended for circuit supplied between {{Not a typo|line}} and neutral, or {{Not a typo|line}} and earthed middle conductor;
:* For residual pulsating direct currents superimposed on smooth direct current.
;'''Type B''' [[file:Typ A-Symbol.png|40px|RCD type B symbol part 1]]&nbsp;[[file:Typ F-Symbol.png|40px|RCD type B symbol part 2]]&nbsp;[[file:Typ B-Symbol.png|40px|RCD type B symbol part 3]]: These trip in the same conditions as Type F, and additionally:
:* For residual sinusoidal alternating currents up to 1{{nbsp}}kHz;
:* For residual alternating currents superimposed on a smooth direct current;
:* For residual pulsating direct currents superimposed on a smooth direct current;
:* For residual pulsating rectified direct current which results from two or more phases; 
:* For residual smooth direct currents, whether suddenly applied or slowly increased, independent of polarity.

The [[BEAMA]] RCD Handbook notes that types F and B have been introduced because some designs of types AC and A can be disabled if a DC current is present that saturates the core of the detector.<ref>BEAMA RCD Handbook - Guide to the Selection and Application of RCDs</ref>

===Directionality===
RCDs may be uni-directional or bi-directional. Bi-directional devices have recently been introduced to address the problem of traditional uni-directional devices being unsuitable for certain configurations of home generation systems ([[Photovoltaic_system|PV]]).

===Surge current resistance===
The surge current refers to the peak current an RCD is designed to withstand using a test impulse of specified characteristics. The IEC 61008 and IEC 61009 standards require that RCDs withstand a 200{{nbsp}}A "ring wave" impulse.   The standards also require RCDs classified as "selective"  to withstand a 3000{{nbsp}}A impulse surge current of specified waveform.