The maintenance manager's guide to circuit protection

Maintenance Manager’s Guide

Power Quality

The Maintenance Manager’s Guide to circuit protection

Hydraulic-magnetic CB operation; a, b, c - overcurrent; d - fault current

Power Quality

Power Quality www.leonardo-energy.org

1. Why is circuit protection installed? The primary objective of Electrical Installation Practice is to provide an installation that is safe and functional. Since it is inevitable that faults will sometimes occur in electrical systems and the appliances that they supply, steps need to be taken to ensure that the safety of people and property is maintained. For the protection of people, exposure to dangerous voltages must be prevented by, for example, good insulation of live parts, proper earthing and earth fault detection. For property protection it is necessary to prevent over-currents that could cause overheating and fire and fault current, i.e. the uncontrolled flow of energy that might lead to ignition or explosion. This document is concerned primarily with protection against the effects of over-currents and fault currents, but, of course, the rapid disconnection of faults enhances safety by reducing the risk of exposure of people to dangerous voltages. Typically, electrical installations follow a tree architecture, the root of which is the point of common coupling (PCC) where there is a protective device provided by the energy supplier. At this point, the supply is defined in terms of capacity – the maximum power that can be normally drawn – and the prospective short circuit current – the maximum short circuit current that could flow through a solid short circuit applied at the PCC. Next in line are the installations main switch and distribution board where the supply splits into a number of sub-circuits, which may be final circuits or feeders to other sub-distribution boards, each with a protective device, as shown in Figure 1.

Figure 1 – Generic installation topology

2

The Maintenance Managers’ Guide to circuit protection www.leonardo-energy.org

2. How are protective devices selected? At every point where the current carrying capacity of the conductor changes (e.g. points a to d), there is a circuit breaker, for which there are three specific requirements: • it must be capable of breaking the maximum prospective fault current

at that point • it must respond to over-current in such a way as to disconnect the

circuit before the excess heat generated in the load circuit cable is sufficient to damage the cable or materials in contact with it • it must limit potential damage to the load circuit by limiting the

magnitude, duration or energy of a fault current to a safe level while disconnecting the circuit from the supply. In an ideal situation, only the breaker for the faulted circuit will open, disconnecting the fault and leaving the rest of the installation unaffected. This is essential for critical loads, but it is often difficult to achieve completely at an affordable cost, so alternative strategies are also used.

3. Prospective fault current The prospective fault current is the maximum current that could flow at a particular point of the installation if a solid short circuit were to be applied there. The prospective fault current depends on the supply impedance at that point, including the source impedance and all cables and accessories in the circuit, so, assuming that there are no transformers, it decreases as electrical distance from the source increases. The prospective fault current is very important in the selection of the protection strategy and of the individual protection devices. Every protective device must be either capable of breaking this current (at its position in the installation) without excessive arcing and without being damaged in the process or must be assisted to do so by an upstream circuit breaker. This is discussed further under ‘Selectivity or discrimination’. 3


4. Circuit breakers There are three main types of circuit breaker used within installations: • Miniature circuit breakers, MCBs, are used in residential, commercial

and industrial applications for final sub-circuit protection. They are relatively cheap, compact and available in a wide range of ratings (5 to 100 A), but have limited breaking capacity, typically 6, 10 or 15 kA. • Moulded case circuit breakers, MCCBs, are designed to high

breaking current capacity with low let-through energy. They are available in frame sizes from 100 A up to about 3000 A for installations where the prospective short circuit fault currents could be as high as 100 kA. The characteristics are not standardised and the trip levels and trip times are often adjustable to provide the desired type and level of discrimination. • Air circuit breakers, ACBs, are more correctly defined as Power

Circuit Breakers, with the fundamental difference being that the short time withstand current of the ACBs is equal to the interrupting rating.

5. Characteristics A typical CB characteristic curve is shown in Figure 2.

Figure 2 – Characteristic curve of a circuit breaker

4


The so-called ‘inverse time’ part of the characteristic is designed to protect against over-current. It allows for substantial short overloads without tripping, because the rate at which the cable conductor temperature rises due to the extra heat generated is relatively slow due to the high specific heat of the copper conductors. As the over-current level increases, the time to respond reduces rapidly to restrict the rise in temperature and reduce the risk of damage. The characteristic takes advantage of the inherent short time over-current tolerance of the cable and allows short duration inrush currents to flow without tripping the breaker. The instantaneous characteristic is intended to respond very rapidly to fault current. Fast action is needed because fault currents are high enough to pose a high risk of damage to load circuits. The most common to achieving these characteristics is the thermalmagnetic breaker. The thermal characteristic is provided passing the load current through an element including a bi-metal strip which deflects according to its temperature. Once a set deflection has been reached, the mechanism is tripped, disconnecting the load circuit. Since the trip is sensitive to temperature, a relatively small over-current will build up heat and eventually cause tripping over an extended time, while a larger overcurrent will heat up and cause tripping in a shorter time. In this area of operation, the device will not respond to very short duration over-currents. The magnetic characteristic is intended to protect against fault currents. It is provided by a small solenoid which exerts a force on the release mechanism. At a predetermined multiple of the rated current, the force is sufficient to operate the trip mechanism, and the load circuit is disconnected. Breaking fault current rapidly is not easy. Without special measures, an arc will form as the contacts separate which will be sustained until the next current zero crossing point. It is possible that a half cycle of fault current will flow, feeding a relatively large amount of energy into the fault. Reducing the energy supplied to the fault requires rapid opening of the contacts and fast quenching of the arc.

5


To ensure rapid opening, the contacts are designed (Figure 3) so that the magnetic force generated by the current flowing up one contact and down the other tends to push the contacts apart.

Figure 3 MCB construction

As the contacts part, an arc is formed and current continues to flow. The arc suppressor, a stack of U-shaped steel plates forming a channel around the contacts, extinguishes the arc. The magnetic field produced by the arc forces the ionised gasses into the plates, rapidly cooling and dividing the gasses and so breaking the arc.

Figure 4 – Arc quenching during MCB operation

6


An alternative method to achieving the inverse-time operating characteristic in circuit breakers, is to use a hydraulic-magnetic principle. The device consists of a solenoid with a moving core which is normally displaced by a spring so that it is outside the magnetic circuit. The core is sealed within a cylinder filled with viscous silicone oil so that its movement is rate limited or ‘damped’. The moving armature is coupled to the tripping mechanism such that when the armature is attracted to the solenoid pole piece, the breaker will trip. Under normal load current the magnetic force generated by the air cored coil is insufficient to overcome that exerted by the spring, so no movement takes place (a). Upon the occurrence of an over-current, the magnetic force induced in the coil exceeds that of the opposing spring. The magnetic core moves towards the pole piece at a rate determined by magnetic force, the viscosity of the silicone oil and the mechanical clearance between the magnetic core and the enclosing tube (b). In due course, the magnetic circuit is completed and the armature is attracted, tripping the circuit breaker (c). The actual time-current characteristic in such devices is easily controllable through a combination of the opposing spring force and the viscosity of the silicone oil that is sealed inside the tube assembly. At very high currents, such as a fault current, the magnetic field generated by the air-cored coil is sufficiently strong to attract the moving armature without the core being in the energised position, so the breaker trips instantaneously (d).

Figure 5 – Hydraulic-magnetic CB operation; a, b, c - overcurrent; d - fault current 7


Electronic sensing for over-current and fault circuit protection in circuit breakers is generally restricted to higher current devices for the present, mainly due to cost considerations. On the other hand, electronic sensing has resulted in a step-function improvement in both the reliability and performance of sensitive earth leakage protection, whilst costs have been held within affordable limits. The use of electronic sensing allows much more accurate protection and enables developments such as annunciators, true RMS protection and communication between circuit breakers. Once cost restrictions have been overcome, technologies such as load and fault signature recognition could become a reality. Intelligent circuit protection is likely to have applications in future ‘Smart homes’.

6. Practical Characteristics In practice, as with any other manufactured device, the performance characteristics of circuit breakers are subject to variation. For MCBs, the thermal characteristic – which provides protection against over-current – the following test points are given in Standards EN 60898: Current I (A)

Nominal trip current

Result

I = 1.13 x In

In (A) All

Must not trip within 1 hour

I = 1.45 x In

< 63A

Must trip within 1 hour

I = 1.45 x In

> 63 A

Must trip within 2 hours

I = 2.55 x In

< 32A

Must trip between 1 and 60 seconds

I = 2.55 x In

> 32A

Must trip between 1 and 120 seconds

This relates to practice as follows. The circuit is designed to provide power to a load or group of loads and so has an expected maximum current, Ib. The nominal MCB rating, In, must be greater than Ib. The conductor size for the circuit is selected to have a current carrying capacity, Iz, that is greater than In. In addition, the current that causes effective operation of the device within the required time, I2, must not be greater than 1.45 times the current carrying capacity of any part of the 8


circuit. As can be seen from the table, for an MCB meeting EN 60898, a current of 1.45 x In will trip in less than one or two hours, so this condition is deemed to be satisfied. The magnetic trip provides protection against the effects of fault current. Since it operates within one cycle, MCBs are sensitive to inrush, starting and surge currents. To avoid a high level of nuisance tripping, MCBs are available with nominal magnetic trip ratings of 5, 10 and 20 times the trip rating as shown in Figure 2. Fault current protection must operate within a prescribed time, which, for 230 V circuits, is 0.4 seconds. This places another requirement on conductor sizing – a short circuit at the far end of the circuit must cause a fault current large enough to operate the magnetic trip. Consequently, there is a maximum limit on the circuit loop impedance according to the class of the breaker. So, if Class D MCBs are used (to avoid nuisance tripping on inrush currents, for example), the loop impedance must be lower than if a Class B device is used. Where the load is concentrated, the loop impedance required by a Class D device should be met if the circuit voltage drop has been correctly taken into account when sizing the cable. However, this may not be the case for distributed loads so it should always be checked. Figure 6 shows the overall responses of Class B, C and D MCBs.

Figure 6 – MCB Characteristics, showing minimum and maximum instantaneous tripping currents 9


Since circuit breakers are thermally operated devices, they are sensitive to temperature with the nominal trip current decreasing as the ambient temperature rises. Devices are designed and tested to operate singly in a vertical orientation at 30° C. At higher temperatures or where the circuit breakers are mounted in groups (as is typical in a distribution board), derating factors must be applied. It must be remembered that circuit breakers also generate heat due to their internal resistance. Manufacturers publish power dissipation values and derating tables for ambient temperatures above 30° C and for grouping factors. As can be seen from Figure 6, the tolerances on circuit breaker characteristics are rather wide. In overload conditions, for example, the time to trip at 1.5 x In is a minimum of 40 seconds and a maximum of 400 seconds – at 30° C. In fault conditions, a Class C device may trip at a minimum current of 5 x In or may not trip until the current exceeds 10 x In. These tolerances must be taken into account in design by using the worst case. For example, when considering the required value of loop impedance, the highest value of tripping current must be used, but when considering resilience to inrush current, the lowest value is the relevant one.

7. Selectivity or Discrimination Ideally, when a fault occurs, only the circuit breaker immediately upstream of the fault should open, thereby isolating the fault without disconnecting any other circuit. In Figure 7, the fault should result in only the breaker C3 opening. In order to achieve that, the discrimination between the breakers at levels A, B and C must be total and breaker C3 must be capable of breaking the prospective fault current at C3. Where resilience is important, total discrimination is essential. In other circumstances it is often necessary to compromise and adopt a less stringent strategy such as ‘backup protection’.

10


Figure 7 – Typical system topology

Discrimination between breakers in a system can be achieved either on the basis of current difference or time difference. A co-ordination study must be undertaken to ensure predictable behaviour. Current discrimination is achieved if the downstream device has a lower current trip level, under all circumstances, than the upstream device. Figure 8 shows an example of this. Since there is no overlap, the downstream breaker will always disconnect the fault and perfect discrimination is achieved. The difficulty is that the downstream breaker must be capable of breaking the full prospective short circuit current, which will often mean that a more expensive device must be used.

11


Figure 8 – Breaker combination giving complete current discrimination

Figure 9 – Breaker combination giving limited current discrimination

8. Cascade or Backup protection Cascade protection is intended to allow the use of lower cost circuit breakers in positions where their current breaking capacity might be less than the prospective fault current. Figure 9 illustrates a case where the breaker characteristics are allowed to overlap at a high fault current level. 12


Here, the upstream breaker will operate first if the fault current exceeds the current at which the curves cross – about 2700 A in this example – otherwise the downstream breaker will operate. In practice this means that fault currents less than 2700 A will be disconnected by the downstream device while larger fault currents, which might exceed the breaking capacity of the downstream device, are disconnected by the upstream device. Many faults will result in a current considerably less than the full prospective fault current, perhaps because the fault has some resistance, or because it occurs (and operates the breaker) at less than full voltage, and will be cleared solely by the downstream breaker. Note that the characteristics in the overload area do not overlap so, for overload conditions, the downstream breaker will always be responsible for clearing overload currents. The disadvantage of this approach is that a high fault current will cause the upstream breaker to operate, removing power from healthy circuits and increasing business disruption. Breaker manufacturers publish comprehensive performance data for cascade protection systems. However, it must be remembered that cascade protection always introduces the probability that power will be removes from a much greater part of the installation than is strictly necessary. Time discrimination is achieved by delaying the action of the upstream breaker until the downstream breaker has had time to open, as shown in Figure 10. This scheme requires the use of suitable breakers; the upstream breaker must be designed for this purpose and the downstream breaker must be capable of breaking the full prospective fault current.

13


Figure 10 – Delaying the operation of the upstream breaker to achieve complete discrimination

Energy selectivity is a relatively modern concept. Since circuit breakers are current and time sensitive, it follows that, for the fault current region, there is a maximum amount of energy that can flow through the breaker before it trips. This is referred to as the ‘energy let-through’ and is expressed in kA2t. Energy discrimination is achieved if the downstream breaker has a lower energy let-through than the upstream breaker. Wiring codes recommend that specific testing should be carried out to verify that discrimination is achieved while manufacturers publish letthrough data on all their devices.

9. What issues are likely to arise in maintenance? 9.1. ‘Early-life’ problems Often new installations experience numerous random protection issues, usually in the form of nuisance tripping or lack of discrimination. These problems can often be traced to incomplete commissioning, where adjustable breakers have been specified by the designers, but the 14


prescribed adjustments have not been correctly carried out – often installed breakers are found in the ‘as shipped’ state. It is important that these problems are rectified by reference to the design documentation rather than trial and error – the erroneous operation may be delayed, but it may also be catastrophic. 9.2. Post commissioning issues and modifications The most common problems are, as for new installations, nuisance tripping and a perceived lack of discrimination, but the cause is different – the use of the installation has changed. Nuisance tripping usually becomes apparent following a change in the nature, number or use of the loads connected to the network. In one case, a tutorial room in an academic institution had been converted into a small computer suite housing 42 personal computers. For software management purposes, these computers were audited and updated overnight, using ‘Wake on Lan’ or ‘magic packet’ instructions to bring them out of standby and into full power mode. These instructions were sent in rapid succession, resulting in a virtually simultaneous power-up of all the computers. The result was a large inrush current, retrospectively measured at >500 A in the worst case, resulting in the MCBs operating. The ‘corrective action’ taken was to replace the Class B MCBs – which trip at 5 times their nominal rating – with Class D devices that would trip at 20 times their nominal rating. This would have been a reasonable response, if the loop impedance had been checked to ensure that a true fault current would have been high enough to trip a Class D breaker. No such check had been made, so the nuisance tripping issue was resolved, but the circuit may not be protected in the event of a real fault! The moral is that any maintenance action that could alter the behaviour of the protection system should be carefully checked. Where there is a perceived lack of discrimination it is usually brought about by a change of function in that area of the installation, such as the installation of more mission critical equipment, for example, or simply a change in the perception of the criticality of the existing equipment. Complete discrimination schemes can be more costly and may have been implemented only in areas where they were considered essential for 15


critical operations. When changes of use occur, it is possible that the protection scheme will need to be readjusted or upgraded.

10. Conclusions The design of the protection scheme requires a systems approach. It should not be assumed that additional circuits can be added to a distribution board without some consideration of the possible effects on upstream protection devices and their settings. Any change of wiring that could affect the loop impedance – such as conductor upsizing or re-routing via a longer or shorter path – must be considered in conjunction with the effect it could have on prospective fault currents and on clearing times for up and downstream protection devices. Even the apparently innocuous change of substituting, say, a Class D MCB in place of a Class B MCB requires reconsideration of loop impedance.

16

The maintenance manager's guide to circuit protection

Recommend Documents