Ring circuit

From DIYWiki
Jump to navigation Jump to search
Diagram of a possible configuration of ring final circuit. Consumer unit is at bottom left.

A ring final circuit or ring circuit (informally also ring main or just ring) is a wiring technique developed and used in the UK that uses two conductors for each of live, neutral and earth to supply each connected socket (or other load).

This design reduces the consequences of faults, and enables the use of smaller-diameter wire than would be used in a radial circuit of equivalent ampacity.

If the load is evenly split across the two halves of the ring, the current in each direction is half of the total, allowing the use of cable with half the current-carrying capacity. In practice, the load is not this evenly split, so the cable used is somewhere between half and full circuit rated current.


The ring circuit cable starts at the consumer unit (fusebox]), visits each socket in turn, and then returns to the consumer unit. The 2 cable ends at the CU are connected together in all modern rings, and supplied by one fuse or MCB.

Ring circuits are widely used to supply sockets for 13A plugs. They are generally wired with 2.5 mm² cable and protected by a 30A fuse, an older 30A circuit breaker, or a 32A circuit breaker. Sometimes 4 mm² cable is used if very long cable runs (causing volt drop issues) or derating factors such as thermal insulation are involved. 1.5 mm² MICC cable ('pyro') may also be used (MICC cable can withstand higher temperatures than PVC) if voltage drop is within spec.

Lay people sometimes refer to any circuit as a "ring" and the term "lighting ring" is often heard from novices. It is not unheard of to see lighting circuits wired as rings (though usually still with a breaker below the cable rating) in DIY installations.

History and use

The ring circuit and BS 1363 plug and socket system were developed in Britiain during 1942-1947.(ref 1) They are commonly used in the United Kingdom, and to a lesser extent in the Republic of Ireland.

The ring circuit came about because Britain had to embark on a massive rebuilding programme following World War II.(ref 2). There was an acute shortage of copper, and it was necessary to come up with a scheme that used less copper than would normally be the case. The scheme was specified to use socket outlets with 13 Amp fused plugs.

Several designs for the plugs and sockets appeared. Only the square pin BS1363 system survives today, but the round pin D&S system was still in use in many locations well into the 1980s. This latter plug had the distinctive feature that the fuse was also the live pin and unscrewed from the plug body.

The ring circuit was devised during a time of copper shortage to allow two 3kW heaters to be used in any two locations and to allow some power to small appliances, and to keep total copper use low. It has stayed the most common circuit configuration in the UK, although the 20A radial (essentially breaking each ring in half and putting the halves on a separate MCB) is becoming more common. Splitting a ring into two 20A radials can be a useful technique where one leg of the ring is damaged and can't be easily replaced.

Another advantage of ring circuits in their early days was an economy of cable and labour, due to the fact that one could simply connect a cable between two existing 15A radially wired sockets to make one 30A ring, then adding as many sockets as were desired. This was an important consideration in the austerity of the 1940s. This left the ring supplied by 2x 15A fuses, which worked well enough in practice.

Many pre-war (round pin) installations used double pole fusing. When 2x 15A radials were converted to a ring on these systems, the ring would then be supplied by no less than 4 fuses! Its very rare to find such circuits still in service today.

Installation rules

Historically the Regs applying to ring circuits said the cable rating must be no less than two thirds of the rating of the protective device, which is 21.3A in the case of a circuit with 32A MCB. An ammendment to the regs in in 2002 changed this value to 20A. This means that the risk of sustained overloading of the cable can be considered minimal. In practice, however, it is uncommon to encounter a ring with a protective device other than a 30A fuse, 30A breaker or 32A breaker, and a cable size other than those mentioned above.

The IEE Wiring Regulations (BS 7671) permit an unlimited number of socket outlets to be installed on a ring circuit, provided that the floor area served does not exceed 100 m². In practice most small and medium houses have one ring circuit per storey, with larger premises having more.

Additional circuits are often fitted for areas of high demand. For example it is common practice to put kitchens on their own ring circuit, or sometimes a ring circuit shared with a utility room, to avoid putting a heavy load at one point on the main downstairs ring circuit. A heavy concentration of load close together on a ring circuit can cause some short term overcurrent in one of the cables if near the end of the ring, so kitchens should not be wired at one end of a ring circuit.

Unfused spurs from a ring wired in the same cable as the ring are allowed to run one single or double socket or one fused connection unit (FCU). The use of two singles was previously allowed, but was banned because people replaced them with doubles. Spurs usually begin from a socket, but are sometimes joined to the ring cable with a junction box. Triple and larger sockets are generally 13A fused, and therefore can also be placed on a spur.

It is not permitted to have more spurs than sockets on the ring, and it is considered bad practice by most electricians to have spurs in a new installation (some think they are bad practice in all cases).

Where loads other than BS 1363 sockets are connected to a ring circuit, or its desired to place more than one socket for low power equipment on a spur, a BS 1363 fused connection unit (FCU) is used. In the case of fixed appliances this will be a switched fused connection unit (SFCU) to provide a point of isolation for the appliance, but in other cases such as feeding multiple lighting points or multiple sockets, an unswitched one is often preferable. (Putting lighting on a ring is generally considered bad practice in new installation, but is often done when adding lights to an existing property.)

Fixed appliances with a power rating over 3 kW (eg, showers, some electric cookers) or with a non-trivial power demand for long periods (eg Immersion Heaters) are no longer recommended to be connected to a ring circuit, but instead are connected to their own dedicated circuit. There are however plenty of older installations with such loads on a ring circuit.

Maximum cable lengths

This table is a basic DIYers guide to the maximum length of cable allowed when installing a final ring circuit. It assumes that

  • The maximum demand of the circuit (allowing for diversity) will not exceed the rating of the protective device
  • The known and assumed loads are balanced around the circuit
  • A maximum voltage drop of 5%
  • A disconnection time in the event of a fault of 0.4 seconds
  • The cable is not grouped with other circuits that may affect the cable's current carrying capacity.
Protective Device Cable Size Installation Method Maximum length (m)
Rating (A) Type L+N/CPC mm² Reference Method TN-S
Zs <= 0.8 Ω
Zs <= 0.35 Ω
30 BS 1361 2.5/1.5 A,100,102,C 111 59zs 111 111
BS 3036 111 49zs 111 111
32 BS 88-2.2, BS 88-6 2.5/1.5 A,100,102,C 106 41zs 106 106
B Type MCB 106 106 106 106
C Type MCB NPsc NPzs 82sc 63zs
30 BS 1361 4.0/1.5 A,100,101,102,C 183 69zs 183 159zs
BS 3036 183 57zs 183 147zs
32 BS 88-2.2, BS 88-6 4.0/1.5 A,100,101,102,C 176 47zs 176 137zs
B Type MCB 176 127zs 176 176
C Type MCB NPsc NPzs 133sc 73zs
  • NP = not permitted
  • zs = limited by earth fault loop impedance
  • sc = limited by neutral loop impedance
  • Any installation method less onerous that that specified may be used
  • Allowed Installation Method info


The final ring-circuit concept has been criticized in a number of ways, and some of these criticisms could explain the lack of widespread adoption outside the UK.

The only way to see the pros and cons of ring circuits is to compare them to the other option, radials.

Fault conditions are not apparent when in use

Ring circuits continue to operate without the user being aware of any problem if there are fault conditions or installation errors that make the circuit unsafe. (ref 3) (ref 4)

  • Part of the ring missing or loose connections result in 2.5 mm² cables running above rated current at times, resulting in reduced cable life. (ref 5)
    • Radials with a loose connection will overheat severely and be an immediate fire risk
    • Radials with a broken connection will not function (if L or N broken), or function with no safety earth connection (if E broken).
  • Accidental cross connection between two 32 A rings means that the fault current protection reaches 64 A and the required fault disconnection times are violated grossly
    • Testing at installation addresses this.
  • Ring spur installations encourage using three connectors in one terminal, which can cause one to become loose and overheat
    • The same is true with both radial and ring circuits when branching off is used.
  • Rings encourage the installation of too many spurs on a ring, leading to a risk of overheating, especially if spur cables are too long
    • Only true if spur cable is underrated. A spur or branch performs exactly the same whether connected to a ring or radial circuit.

Complexity of safety tests

Testing existing ring circuits where the true topology is unknown, and inadvised modifications may have been made to the circuit can take 5-6 times longer than testing a radial circuit where the topology is basically not relevant.(ref 4)

The installation tests required for a new ring are are in many cases slightly simpler than those required for a radial socket circuits.

Balancing requirement

Regulation 433-02-04 of BS 7671 requires that the installed load is distributed around the ring such that no part of the cable exceeds its capacity. This requirement is difficult to fulfill and may be largely ignored in practice, as loads are often co-located (washing machine, tumble dryer, dish washer all next to kitchen sink) and not necessarily near the centre of the ring.(ref 4)

Electromagnetic interference

Ring circuits can generate magnetic fields. In a normal (non-ring, radial) circuit, the current flowing in the circuit must return through (almost exactly) the same path through which it came, especially if the live and neutral conductors are kept in close proximity to each other. This prevents the circuit forming a large magnetic coil (loop antenna), which would otherwise induce a magnetic field at the AC frequency (50 or 60 Hz). In a ring circuit, on the other hand, it is possible that the live and neutral currents are not equal on each side of the ring. Mains-frequency currents follow the path of least resistance, and it is possible, especially with aging oxidized contacts (note this is a fault condition, since all connections should be gas-tight) that from a socket, the lowest-resistance live connection is along the left-hand side of the ring, and the lowest-resistance neutral connection is along the right-hand side. As a result, current is flowing around the ring and will therefore induce a magnetic field. In the extreme case of a defective ring circuit, the live connection could become completely interrupted on one side of the ring and the neutral connection on the other, and then the full current would supply the magnetic field. This can lead to interference such as such as mains hum in poorly substandard audio devices.

On the other hand high resistance connections on radial circuits result in overheating and fire risk, a far more serious problem.

Overcurrent protection

Ring circuits are general purpose circuits, and hence the main protective device (fuse or MCB) must provide protection against both fault currents and overload current as far as each socket outlet. However the plug fuse provides the fault current protection for the appliance flex (if an appliance also requires overload protection, then it must include its own fuse or other protective device). Since the circuit protective device does not need to protect the appliance flex from fault currents, the circuit can be protected with a high-rated overcurrent device (typically 30/32 A) appropriate for the protection of the pair of 2.5mm² T&E cables. This allows ring circuits to supply a large number of sockets with a substantial overall capacity of 7.2kW, without the user of the circuit needing to pay particular attention to where high current appliance loads are connected. Adding greatly to the flexibility and versatility of the circuit (the same logic applies to 32A radial circuits, however the physical characteristics of the 4mm² cable required makes them far more difficult to wire in practice.)

Fused plugs

The circuits used in most other countries don't have the benefit of fused plugs, and hence the main circuit breaker must also provide fault current protection as far as each appliance, including its flex. Hence the maximum current for the whole circuit needs to be restricted, often to 16A in order to maintain fault current protection for the appliance flexes. This is turn restricts the supply to a small number of sockets for each circuit.

This incompatibility in the fault current protection of appliance flexes with the high current ring and radial circuits popular in the UK, and the radials with unfused plugs favoured in many other countries, makes the UK system difficult to adopt in other countries, and in turn this has been another major stumbling block on the road to worldwide standardization of domestic AC power plugs and sockets.

Note that although plug-fuses can, in principle, be better matched to the maximum current required by an appliance, in practice, some plugs in the UK are merely fitted with a fuse of the maximum permitted rating of 13A. This is not a problem since all appliance flexes are required to be safe with a 13A fuse, but it does mean the potential safety advantage is only partially realised. The introduction of regulations requiring new appliances to be sold with correctly fused pre-fitted plugs improved this situation.

There is an addition benefit to the plug fuse in that an appliance with a blown plug fuse will not be live when plugged in again, whereas with fuseless plugs, a faulty appliance may remain dangerous to plug in, and another person will often do so at a later date.


1. Malcolm Mullins: The origin of the BS 1363 plug and socket outlet system. IEE Wiring Matters, Spring 2006.

2. D.W.M. Latimer: History of the 13 amp plug and the ring circuit. Presentation papers from a public meeting to discuss the issue of ring circuits, IET, London, October 2007 (PDF in ZIP)

3 Roger Lovegrove: Ringing the changes. EMC, April 2006

4 Roger Lovegrove: Ring circuits – the disadvantages. Presentation papers from a public meeting to discuss the issue of ring circuits, IET, London, October 2007 (PDF in ZIP)

5. P Knowles: Ring main lining. EMC, February 2007

See also