![Publication No Cu0141
Issue Date: August 2015
Page 2
FUNDAMENTALS OF SURGE PROTECTION
GENERAL
In the low-voltage (end-user) environment, SPDs serve to divert impinging surges by offering a low-impedance
path to return the surge current to its source. This function can be accomplished in a single or several stages,
depending on the system configuration and the degree of freedom available to the user in connecting SPDs at
different points of the system.
Large surges originating outside of the user's facility – usually associated with lightning or major power-system
events – are best diverted at the service entrance of a facility. Surges generated within the premises can be
diverted by SPDs located close to the internal surge source or close to the sensitive equipment in need of
protection. Figure 1 shows the principle of a two-stage protection scheme. The first stage provides diversion of
impinging high-energy surges
*
through the arrester, typically installed at the service entrance, or by a device
permanently connected at the service panel. Some restriction of the propagation of surge currents in branch
circuits is inherently provided by the inductance of the premises wiring. The second stage of voltage limiting is
provided by an SPD of lesser surge-handling capability which is typically located close to the equipment in
need of protection as an add-on, plug-in device, or incorporated within the equipment by the manufacturer.
This second stage completes the scheme for dealing with surges of external origin as well as for surges
originating within the building.
Figure 1 – A two stage protection scheme.
*
The term high-energy surge is used here as short-hand for the more accurate term surge with high energy-
delivery capability. Some power-quality surveys reports make reference to the concept of and term energy in
the surge as an abstract characteristic of the surge itself, which could be derived from measuring only the
voltage and duration of a surge occurring in the power system Such characterization is meaningless and
misleading (Standler, 1989); (Key et al., 1996].](https://image.slidesharecdn.com/cu0141antransientsandovervmitigationv2-190712170451/85/Transient-overvoltages-and-currents-mitigation-and-protection-techniques-5-320.jpg)
![Publication No Cu0141
Issue Date: August 2015
Page 7
SELECTION OF LIMITING LEVELS
Under the perception that it is necessary – or at least desirable – to provide the lowest possible level for the
limiting (clamping) voltage, some SPD manufacturers have been offering SPDs with a limiting level as low as
330 V for 120 V circuits. These manufacturers claim superior performance over competing SPDs that achieve
somewhat higher levels (such as 400 V or 500 V) under the same specified applied surge current. A similar
downward auction for limiting voltage selection is likely to occur for other system voltages.
During the development of the seminal standard UL 1449 and IEC Publication 664 in the late seventies and
early eighties, a table of overvoltage categories was under consideration with levels that included the 330 V
value. However, it is noteworthy that this level was not cited in IEC 664 as a level for systems rated at 120 V,
but rather as a level for systems rated at 50 V. Thus, the understandable desirability of effective protection
through low limiting voltage became biased toward considering (advertising) a limiting voltage of 330 V as a
plus factor in evaluating the performance of SPDs for 120 V circuit applications.
A better perspective of the issue was reached in the nineties, but the outmoded perceptions and advertising
claims linger. Against these, the following facts and considerations should be kept in mind:
- Among published papers on immunity testing of appliances [Anderson & Bowes, 1990, Smith &
Standler, 1992], and surge immunity tests performed on a wide range of appliances, no adverse
effects have been reported until the applied surges exceed thresholds of 600 V to 1,000 V
- The well-accepted CBEMA Curve (now renamed ITIC Curve) did not require limiting voltages of 330 V
(for 120 V loads) for durations of less than 0.5 ms, and allowed limiting voltages with increasing levels
at shorter durations
- A low limiting level can result in premature aging of the SPD when it is called upon to more frequently
clamp surges that would not have caused any intervention of an SPD with somewhat higher limiting
voltage [Martzloff & Leedy, 1989], or swells resulting in significant shift of the nominal voltage when
applied in large numbers over the life of the device [Lagergren et al., 1992]
- Some utilities wish to select an SPD capable of withstanding twice the line-to-neutral voltage in order
to avoid device failure or fuse opening under the condition of lost neutral connection
It appears then that the advantage of a low limiting voltage can in fact have undesirable side effects on the
long-term overall reliability of the application of an SPD in end-user circuits. The chart in Figure 4 summarizes
the considerations involved in selecting a suitable SPD. Clearly, the mere selection of a limiting level as a
retrofit approach to protect a piece of equipment is insufficient for a reliable application.](https://image.slidesharecdn.com/cu0141antransientsandovervmitigationv2-190712170451/85/Transient-overvoltages-and-currents-mitigation-and-protection-techniques-10-320.jpg)
![Publication No Cu0141
Issue Date: August 2015
Page 12
After all, the presence of a reactor connected in series with a compensation capacitor should be common
practice today. Such a combination will act as a detuning reactor to avoid resonances between the
compensation capacitance and any inherent inductance in the system. Such resonances could be excited by
lower order harmonics present in the system. The inductance ratings of such detuning reactors, however, are
one or two orders of magnitudes higher than of those for mitigating the inrush currents. Consequently, the
current transient when connecting a capacitance is replaced by a voltage transient when disconnecting an
inductance. This needs to be dealt with accordingly.
MOV ARRESTERS
Metal oxide varistors (MOVs) can limit transient overvoltages to the protective level (typically 1.8 to 2.5 per-
unit) at the point of application. The primary concern associated with MOV application is the energy duty
during a restrike event. Although a rare occurrence, a switch restrike generally results in the highest arrester
duty for arresters located at the switched capacitor. In addition, remote arresters (including low-voltage
customer applications) can be subjected to severe energy duty if voltage magnification occurs. This condition
could be especially troublesome for distribution systems if SiC-based arresters remain in service.
The effectiveness of these mitigation methods is system dependent, and a detailed analysis is necessary to
select the optimum scheme [Mikhail et al., 1986]; [Grebe, 1995]. While often justifiable for large transmission
applications, transient analysis of distribution capacitor applications is rarely performed and, in general,
capacitor banks are installed without transient overvoltage control.
If these preventive measures against capacitor switching transients are not practical or possible for an end-
user customer to coordinate with the supplier, then remedial measures can be applied at the service entrance
and points of connection of equipment in the customer's facility. This is described in the subchapter Function
of SPDs.](https://image.slidesharecdn.com/cu0141antransientsandovervmitigationv2-190712170451/85/Transient-overvoltages-and-currents-mitigation-and-protection-techniques-15-320.jpg)
![Publication No Cu0141
Issue Date: August 2015
Page 15
Figure 6
Coordination between two SPDs can be demonstrated by five different approaches, described below.
Whatever the approach, it is necessary that the end-user have control over the selection of the SPDs,
preferably both of them. If one is already given and beyond the control of the end-user, the coordination of
the second candidate SPD can be assessed and more suitable candidates identified if necessary. Most of the
time, coordination is, or might seem to be a complex situation. Some of the following five procedures might
then appear to the end-user as difficult to apply, due to lack of knowledge concerning some data, as for
example accurate SPD characteristics. Nevertheless, coordination is necessary to ensure technically effective
and cost-effective use of resources.
ASSESSMENT OF COORDINATION
The five basic different methods of assessing the coordination of two SPDs are:
1. Application of preferred SPD combinations. This is the most convenient variant for the users because
the burden of demonstration is incumbent upon the SPD manufacturer. However, this approach
implies either that a sole source manufacturer be selected, or that standard methods be developed to
allow determination that candidate devices from different sources are in fact equivalent.
2. Coordination computation. Computer simulation enables complex systems to be examined and
parametric evaluation performed over a wide range.
3. Application of the Let-Through Energy concept. In this concept, some decoupling impedance is
postulated in the upstream SPD and a computation is performed for converting the SPD
characteristics on the basis of an equivalent combination wave generator. A comparison can then be
made with the energy-withstand capability of the downstream SPD (Hasse et al., 1994]
4. Coordination test. A full-scale test is performed with candidate SPDs, for a postulated decoupling
impedance and a range of surge currents such that blind spots, if any, are revealed. Candidate devices
may include voltage-limiting SPDs, voltage-switching SPDs, or combination type SPD [voltage-
switching SPD + voltage-limiting SPD].
5. Simplified rules with margin. Simplified tables including margins for the coordination of some typical
SPDs may be used when no other data is known from the SPD manufacturers. The values given in
these tables include sufficient margins to cover discrepancies between manufacturers and
manufacturing tolerances.](https://image.slidesharecdn.com/cu0141antransientsandovervmitigationv2-190712170451/85/Transient-overvoltages-and-currents-mitigation-and-protection-techniques-18-320.jpg)
![Publication No Cu0141
Issue Date: August 2015
Page 16
SYSTEM INTERACTION PROTECTION
As discussed in Cu0138, interactions at equipment ports involving different systems can produce damaging or
upsetting shifts in reference potentials during surge events. Anecdotal information indicates that this
mechanism is probably the most frequent cause of electronic consumer equipment failure (TV receivers and
VCRs with cable signal input, computers with wired modem connection, smart telephones, etc.)
Various mitigation schemes have been proposed to remedy upsetting or damaging potential differences.
Increasing the withstand capability of the PC system by the manufacturers is unlikely to occur, given the
market economics. Moreover, it actually might not be practical for some of the voltages that can appear in
actual situations. The most effective course would be a fibre optic decoupling, inserted in the communications
link, but the expense and involvement required would be objectionable for the typical home office application.
For industrial applications where the consequences can be severe, this approach is often selected and has
been found to be both successful and cost-effective
NON-METALLIC LINKS
Consumer options are limited to either replacement of existing conducted communication and control links
with their fibre optic counterparts, or insertion of commercially available isolation devices as needed. The first
option, while appropriate for new installations, might not be cost-effective for an existing installation. Devices
manufactured specifically for various port types can be applied for existing installations. For example, RS-232
communications are often used for connecting programmable logic controllers and other devices in utility and
industrial facilities. These links are easily isolated using an optically coupled adapter that connects directly to
the RS-232 port. These devices must have a suitable voltage isolation rating (e.g. kilovolts) for the application.
Some of these devices only have isolation capability measured in hundreds of volts which might not be
sufficient. Similar devices exist for other ports.
EQUALIZING REFERENCE POTENTIALS
Theoretically, a simple way to eliminate the shifts in reference potentials during surge events would be to
bond the two reference conductors of the two systems. However, such a simple bonding might not be
acceptable from the point of view of the system operators, the local or national codes, and product listing
agencies. Therefore, this Guide cannot provide recommendations applicable to all situations in all countries,
but only describe general principles leading to a solution.
When bonding the two references directly is not permitted, it might be permissible to provide an SPD of
appropriate characteristics that will in effect bond the two references – although with some potential
difference still left – but only during a surge event. This legitimately circumvents the prohibition of a
permanent bonding under steady-state conditions. The alternative, i.e. no bond at all, is that unintended
bonding is likely to occur – with damaging results – by sparkover of the clearances separating the conductors
of the two systems. For instance, tests have shown that the isolation between the cable input and the tuner of
a typical TV receiver will flashover at about 2.5 kV [Martzloff, 2000].](https://image.slidesharecdn.com/cu0141antransientsandovervmitigationv2-190712170451/85/Transient-overvoltages-and-currents-mitigation-and-protection-techniques-19-320.jpg)
Transient overvoltages and currents: mitigation and protection techniques
- 1. APPLICATION NOTE TRANSIENTS & OVERVOLTAGES: MITIGATION AND PROTECTION TECHNIQUES UIE August 2015 ECI Publication No Cu0141 Available from www.leonardo-energy.org
- 2. Publication No Cu0141 Issue Date: August 2015 Page i Document Issue Control Sheet Document Title: AN – Transients and Overvoltages: Mitigation Publication No: Cu0141 Issue: 03 Release: August 2015 Author(s): UIE Reviewer(s): Roman Targosz, Stefan Fassbinder Document History Issue Date Purpose 1 June 2007 Original publication 2 February 2012 Adaptation for adoption into the GPG 3 August 2015 Review Disclaimer While this publication has been prepared with care, European Copper Institute and other contributors provide no warranty with regards to the content and shall not be liable for any direct, incidental or consequential damages that may result from the use of the information or the data contained. Copyright© European Copper Institute. Reproduction is authorised providing the material is unabridged and the source is acknowledged.
- 3. Publication No Cu0141 Issue Date: August 2015 Page ii CONTENTS Summary ........................................................................................................................................................ 1 Fundamentals of surge protection.................................................................................................................. 2 General ..................................................................................................................................................................2 Types of surge-protective device components.......................................................................................................3 Types of surge-protective devices..........................................................................................................................3 Function of SPDs............................................................................................................................................. 6 Performance under surges .....................................................................................................................................6 Performance under steady-state variations...........................................................................................................6 Durability (Endurance)............................................................................................................................................6 Modes and mode conversion at the power port....................................................................................................6 Selection of limiting levels .............................................................................................................................. 7 Mitigate surges at the origin........................................................................................................................... 9 Mitigation of capacitor-switching transients................................................................................................. 10 General ................................................................................................................................................................10 Timing control.......................................................................................................................................................10 Pre-insertion devices ............................................................................................................................................11 Fixed inductors .....................................................................................................................................................11 MOV arresters ......................................................................................................................................................12 Inherent or built-in equipment protection.................................................................................................... 13 Add-On protection........................................................................................................................................ 14 SPD application at service entrance or point of use.............................................................................................14 Cascade coordination ...........................................................................................................................................14 General objective of coordination .........................................................................................................14 Assessment of coordination...................................................................................................................15 System interaction protection ...................................................................................................................... 16 Non-metallic links.................................................................................................................................................16 Equalizing reference potentials ............................................................................................................................16
- 4. Publication No Cu0141 Issue Date: August 2015 Page 1 SUMMARY This is the eighth publication in a series of Application Notes on transient overvoltages and transient currents in AC power systems and customer installations. For a general introduction to the subject, first read Cu0134 – Transients and Overvoltages: Introduction. This application note provides insight into the fundamentals of surge protection and mitigation. It describes the basic characteristics of Surge Protective Devices (SPDs) which divert surges by offering a low- impedance path to return the surge current to its source. The different types of SPDs are presented, based on their internal circuitry and their components. This application note also discusses the various functions of an SPD and the conditions it should be able to withstand as well as insight into how to select the SPD limiting levels. Different techniques for the mitigation of capacitor-switching transients are discussed in this application note. In the final part of this note, built-in equipment protection, add-on protection, and system interaction protection are discussed in greater detail.
- 5. Publication No Cu0141 Issue Date: August 2015 Page 2 FUNDAMENTALS OF SURGE PROTECTION GENERAL In the low-voltage (end-user) environment, SPDs serve to divert impinging surges by offering a low-impedance path to return the surge current to its source. This function can be accomplished in a single or several stages, depending on the system configuration and the degree of freedom available to the user in connecting SPDs at different points of the system. Large surges originating outside of the user's facility – usually associated with lightning or major power-system events – are best diverted at the service entrance of a facility. Surges generated within the premises can be diverted by SPDs located close to the internal surge source or close to the sensitive equipment in need of protection. Figure 1 shows the principle of a two-stage protection scheme. The first stage provides diversion of impinging high-energy surges * through the arrester, typically installed at the service entrance, or by a device permanently connected at the service panel. Some restriction of the propagation of surge currents in branch circuits is inherently provided by the inductance of the premises wiring. The second stage of voltage limiting is provided by an SPD of lesser surge-handling capability which is typically located close to the equipment in need of protection as an add-on, plug-in device, or incorporated within the equipment by the manufacturer. This second stage completes the scheme for dealing with surges of external origin as well as for surges originating within the building. Figure 1 – A two stage protection scheme. * The term high-energy surge is used here as short-hand for the more accurate term surge with high energy- delivery capability. Some power-quality surveys reports make reference to the concept of and term energy in the surge as an abstract characteristic of the surge itself, which could be derived from measuring only the voltage and duration of a surge occurring in the power system Such characterization is meaningless and misleading (Standler, 1989); (Key et al., 1996].
- 6. Publication No Cu0141 Issue Date: August 2015 Page 3 TYPES OF SURGE-PROTECTIVE DEVICE COMPONENTS Providing detailed information on SPDs is not within the scope of this Guide; however, some knowledge about the components involved in the protective function will assist in making engineering judgements about their selection and application. SPD components can be classified into two types: voltage-limiting devices and voltage-switching devices (also called crowbars). A low-pass filter will attenuate the high frequency portion of a surge and thus act in a manner similar to an SPD. However, according to the formal definition of an SPD, such a low-pass filter is not in fact a proper SPD because it does not include a non-linear device. A voltage-switching (crowbar) device involves a switching action as the voltage across the terminals rises. This switching can be either the breakdrown of a gas between electrodes or the activation of a solid-state device. After its activation, the device presents a low impedance to divert the transient current through some low- impedance path back toward its source. A voltage-limiting device can be based on the technology of avalanche (zener) diodes or varistors. Both avalanche diodes and varistors exhibit a monotonic and non-linear current- voltage characteristic, in contrast with the abrupt change of impedance associated with voltage-switching devices. TYPES OF SURGE-PROTECTIVE DEVICES Several types of internal configuration are found in commercial SPDs. This variety is due to the internal circuitry as well as the SPD components used in the circuits. Recent IEEE and IEC standards have now designated SPDs as being one-port SPD or two-port SPD. These two terms are explained below. One-port SPDs – The protective function of a one-port SPD is provided by a shunt connection (Figure 2) to the circuit to be protected, seen as one or more pairs, such as line-to-neutral, line-to-ground, and neutral-to- ground. Figure 2
- 7. Publication No Cu0141 Issue Date: August 2015 Page 4 The issue of providing protection to one, two, or three of these pairs is discussed later in this application note. For the one-port SPD, the limiting voltage is measured at the terminals, either screw terminals or leads. This configuration opens the possibility of misapplying the device by using long leads for the connection, which results in an inductive voltage being added to the intended limiting voltage of the device. This possibility must be recognized in guidance on SPD applications as well as in details of the lead configuration when performing tests. Two-port SPDs – In two-port SPDs, there is a power input port and a power output port that are distinct. Depending upon the design, some components can be inserted in series between the input and output port. The motivation for this configuration is generally to provide the two-stage protection scheme illustrated in Figure 1a within a single package. Another motivation is to remedy the lead length problem: Some manufacturers offer the so-called in-line configuration for permanently connected SPDs, with four terminals and the requirement that the power flow be through the SPD. Figure 3 shows typical arrangements of two-port SPDs. These configurations make it necessary to design and rate the device for a specific value of the power-frequency current. Figure 3
- 8. Publication No Cu0141 Issue Date: August 2015 Page 5 From the point of view of surge suppression, a two-port SPD is intended for one and only one direction of impinging surges. Typically, the SPD component connected nearest to the input terminals has a greater current-handling capacity than the component connected nearest to the output terminals. This asymmetry must be recognized in the application. Note in Figure 3 that the designation of two-port applies whether the SPD has a three-terminal or a four- terminal configuration. Correct operation of this SPD is based on the following assumptions: 1) The impinging surge comes from the left of the figure 2) The series impedance separating the two SPD components will produce a sufficient voltage drop, added to the limiting voltage of the varistor at right to cause the gas tube at left to sparkover and relieve the varistor from the stress of conducting the tail of the surge Such a design might operate quite well at the maximum stress level, but not at intermediate current levels, or slow-rising surges, for which the voltage drop in the series impedance would be insufficient to produce sparkover of the gas tube. Such a situation is referred to as blind spot and should be recognized when specifying selection or tests for SPDs.
- 9. Publication No Cu0141 Issue Date: August 2015 Page 6 FUNCTION OF SPDS The primary function of an SPD is the diversion of impinging surges away from sensitive loads without causing immediate or long-term failure of the SPD. However, the SPD must also withstand the normal and abnormal variations in steady-state conditions of the power system. This dual requirement focuses on characteristics discussed in the following paragraphs. Particular emphasis must be given to the failure modes of a sound SPD under abnormal system conditions, as well as failure modes under normal system conditions of an SPD, which, for whatever reason, might have reached the end of its useful life. PERFORMANCE UNDER SURGES This characteristic is of course the very purpose of an SPD. Various severity levels of surges should be stipulated for candidate SPDs depending upon the location of the SPD. The 0.5 µs 100 kHz Ring Wave and the 1.2/50 µs 8/20 µs Combination Wave provide basic information on the limiting voltage and energy-handling capability of an SPD for single surges as well as for cumulative effects over the life of the device. The purpose of an SPD is to limit transient overvoltages to values that downstream equipment can withstand without damage, or in certain cases without upset. PERFORMANCE UNDER STEADY-STATE VARIATIONS As discussed in Cu0136, low-voltage AC power systems can experience temporary overvoltages that can stress SPDs. The concept and specification of a maximum continuous operating voltage (MCOV) has been developed to assess the ability of an SPD to withstand such abnormal conditions. Proper selection of an SPD requires reconciliation of the MCOV of the SPD and the various scenarios for the occurrence of temporary overvoltages at the point of interest. DURABILITY (ENDURANCE) An SPD is expected to provide reliable operation for a number of years. This durability is generally not specified by the SPD manufacturer (except, indirectly, by those offering a lifetime guarantee on their package) because of the dominant effect of undefined environmental stresses on the aging process. MODES AND MODE CONVERSION AT THE POWER PORT Depending upon the design and application philosophy of manufacturers, commercial SPDs are offered with surge protection provided between all pairs of conductors or between only some selected pairs. In North American practice where the equipment grounding conductor is bonded to the neutral and local earth at the service entrance, only differential mode surges can emerge from the service panel. However, surge currents associated with a differential mode surge, when flowing in the conductors of a branch circuit, can produce voltage drops along the neutral conductor or the equipment grounding conductor. This results in a significant difference of voltages with respect to other grounded structures (a form of mode conversion). The diversity of approaches must be recognized, and any test schedule must be established, taking into consideration the specific design of the SPD under test.
- 10. Publication No Cu0141 Issue Date: August 2015 Page 7 SELECTION OF LIMITING LEVELS Under the perception that it is necessary – or at least desirable – to provide the lowest possible level for the limiting (clamping) voltage, some SPD manufacturers have been offering SPDs with a limiting level as low as 330 V for 120 V circuits. These manufacturers claim superior performance over competing SPDs that achieve somewhat higher levels (such as 400 V or 500 V) under the same specified applied surge current. A similar downward auction for limiting voltage selection is likely to occur for other system voltages. During the development of the seminal standard UL 1449 and IEC Publication 664 in the late seventies and early eighties, a table of overvoltage categories was under consideration with levels that included the 330 V value. However, it is noteworthy that this level was not cited in IEC 664 as a level for systems rated at 120 V, but rather as a level for systems rated at 50 V. Thus, the understandable desirability of effective protection through low limiting voltage became biased toward considering (advertising) a limiting voltage of 330 V as a plus factor in evaluating the performance of SPDs for 120 V circuit applications. A better perspective of the issue was reached in the nineties, but the outmoded perceptions and advertising claims linger. Against these, the following facts and considerations should be kept in mind: - Among published papers on immunity testing of appliances [Anderson & Bowes, 1990, Smith & Standler, 1992], and surge immunity tests performed on a wide range of appliances, no adverse effects have been reported until the applied surges exceed thresholds of 600 V to 1,000 V - The well-accepted CBEMA Curve (now renamed ITIC Curve) did not require limiting voltages of 330 V (for 120 V loads) for durations of less than 0.5 ms, and allowed limiting voltages with increasing levels at shorter durations - A low limiting level can result in premature aging of the SPD when it is called upon to more frequently clamp surges that would not have caused any intervention of an SPD with somewhat higher limiting voltage [Martzloff & Leedy, 1989], or swells resulting in significant shift of the nominal voltage when applied in large numbers over the life of the device [Lagergren et al., 1992] - Some utilities wish to select an SPD capable of withstanding twice the line-to-neutral voltage in order to avoid device failure or fuse opening under the condition of lost neutral connection It appears then that the advantage of a low limiting voltage can in fact have undesirable side effects on the long-term overall reliability of the application of an SPD in end-user circuits. The chart in Figure 4 summarizes the considerations involved in selecting a suitable SPD. Clearly, the mere selection of a limiting level as a retrofit approach to protect a piece of equipment is insufficient for a reliable application.
- 11. Publication No Cu0141 Issue Date: August 2015 Page 8 Figure 4
- 12. Publication No Cu0141 Issue Date: August 2015 Page 9 MITIGATE SURGES AT THE ORIGIN The classical EMC approach to disturbance mitigation generally recommends mitigation at the origin whenever possible. For instance, IEC 61000-5-2 offers guidelines on earthing and wiring practices that can reduce the coupling of surges between their origin and the circuits of interest. In the case of surges however, all the discussions presented on their origin indicate that such a sensible approach cannot apply to lightning surges once they have been coupled into the power system. Designing a power system for less exposure to lightning surges is not within the prerogatives of the industrial user, and thus is not discussed in this Guide. One possible exception to this situation where the end-user is confronted with surges that cannot be mitigated before they impact the installation is the case of capacitor- switching surges discussed below. For surges such as these, mitigation might be possible at the origin when the parties involved cooperate in finding a solution.
- 13. Publication No Cu0141 Issue Date: August 2015 Page 10 MITIGATION OF CAPACITOR-SWITCHING TRANSIENTS GENERAL Devices for the mitigation of capacitor-switching transients at their point of origin either attempt to minimize the overvoltage or overcurrent at the point of application of the capacitors, or limit (clip) the overvoltage. For the capacitor-switching transients discussed in Cu0136, several solutions or mitigation means can be applied, as follows. Solutions to the excessive secondary overvoltages associated with capacitor switching (described in Cu0136) usually involve one or more of the following: a. Detuning the primary circuit by changing capacitor bank sizes, moving banks, and/or removing banks from service b. Switching large banks in more than one section c. Using an overvoltage control method such as pre-insertion resistors/inductors or synchronous closing control d. Converting low-voltage power factor correction banks into harmonic filters (detuning the secondary circuit – which should be common practice today, but is not always fully implemented) Solutions to the excessive current inrush associated with back-to-back capacitor switching (described in Cu0136) usually involve one or more of the following: a. Adding current-limiting reactors to decrease the peak current and frequency of the oscillatory inrush current b. Adding pre-insertion resistors or inductors to the switching device c. Adding synchronous closing control to the switching device d. Electronic switching, connecting presicely at a phase angle for which the residual voltage of the capacitor equals the instantaneous system voltgage Each of these methods has various advantages and disadvantages in terms of transient overvoltage mitigation, cost, installation requirements, operating and maintenance requirements, and reliability. Some of these factors are described in the following paragraphs. TIMING CONTROL Synchronous closing consists of performing the closing of the contacts of each phase near a zero voltage, as illustrated in Figure 5. To accomplish closing at or near a zero voltage, it is necessary to use a switching device that maintains a dielectric strength sufficient until its contacts touch.
- 14. Publication No Cu0141 Issue Date: August 2015 Page 11 Figure 5 Although this level of precision is difficult to achieve, closing consistency of ±0.5 ms should be possible. A closing consistency of ±1.0 ms was found to provide overvoltage control comparable to properly sized pre- insertion inductors. The success of a synchronous closing scheme is often determined by the ability to repeat the process under various system and climate conditions. Adaptive, microprocessor-based control schemes that have the ability to learn from previous events address this concern. Grounded capacitor banks are controlled by closing the three phases at three successive zeros of the phase-to- ground voltage (60 degree separation). Ungrounded capacitor banks are controlled by closing the first two phases at a phase-to-phase zero voltage and then delaying the third phase 90 degrees (zero voltage of phase- to-ground). PRE-INSERTION DEVICES A pre-insertion impedance (resistor or inductor) provides a means for reducing the transient currents and voltages associated with the energization of a capacitor bank. The impedance is by-passed shortly after the initial transient dissipates, thereby producing a second transient event. The insertion transient typically lasts for less than one cycle of the frequency. The performance of the system pre-insertion impedance is evaluated using magnitudes of both the insertion and the by-pass transients, as well as the capability to dissipate the energy associated with the event, and repeat the event on a regular basis. The optimum resistor value for controlling capacitor-switching transients depends primarily upon the capacitor size and source strength. FIXED INDUCTORS Fixed inductors have been used successfully to limit inrush current during back-to-back switching. The value of these inductors is typically on the order of several hundred microhenries. In addition, inductors provided to limit outrush current (into a nearby fault) may be applied, with typical values of 0.5 mH to 2.0 mH. However, these fixed reactors do not provide appreciable transient overvoltage reduction.
- 15. Publication No Cu0141 Issue Date: August 2015 Page 12 After all, the presence of a reactor connected in series with a compensation capacitor should be common practice today. Such a combination will act as a detuning reactor to avoid resonances between the compensation capacitance and any inherent inductance in the system. Such resonances could be excited by lower order harmonics present in the system. The inductance ratings of such detuning reactors, however, are one or two orders of magnitudes higher than of those for mitigating the inrush currents. Consequently, the current transient when connecting a capacitance is replaced by a voltage transient when disconnecting an inductance. This needs to be dealt with accordingly. MOV ARRESTERS Metal oxide varistors (MOVs) can limit transient overvoltages to the protective level (typically 1.8 to 2.5 per- unit) at the point of application. The primary concern associated with MOV application is the energy duty during a restrike event. Although a rare occurrence, a switch restrike generally results in the highest arrester duty for arresters located at the switched capacitor. In addition, remote arresters (including low-voltage customer applications) can be subjected to severe energy duty if voltage magnification occurs. This condition could be especially troublesome for distribution systems if SiC-based arresters remain in service. The effectiveness of these mitigation methods is system dependent, and a detailed analysis is necessary to select the optimum scheme [Mikhail et al., 1986]; [Grebe, 1995]. While often justifiable for large transmission applications, transient analysis of distribution capacitor applications is rarely performed and, in general, capacitor banks are installed without transient overvoltage control. If these preventive measures against capacitor switching transients are not practical or possible for an end- user customer to coordinate with the supplier, then remedial measures can be applied at the service entrance and points of connection of equipment in the customer's facility. This is described in the subchapter Function of SPDs.
- 16. Publication No Cu0141 Issue Date: August 2015 Page 13 INHERENT OR BUILT-IN EQUIPMENT PROTECTION To take care of load equipment performance, well-considered specifications for the equipment should always be presented to the potential suppliers. Inquiries should be carried out with several suppliers to obtain comparable data for equipment performance. This should include levels of immunity to transients, among other data. However, different protection strategies can make protection levels difficult to compare. Solutions for protection proposed by different suppliers should be evaluated. It should be verified whether they are using SPDs internally or base their protection strategy on providing SPDs centralized in the electrical installation. In any event, the supplier should commit to the specified level of transient voltage immunity. General information included in equipment specification should be the conditions defined by mutual agreement among the parties involved. Factors to be considered include likelihood and frequency of occurrence, magnitude, and other characteristics of the transients. Specific equipment performance parameters to be specified may, among others, include the magnitude of standardized transients that should not cause an abnormal operation of the equipment and the level at which the equipment should not be damaged. The expected outcomes for various scenarios should be specified. For instance, the IEC 61000-4-xx series describes the level of performance expected and should be a valuable fundamental reference and support for determining equipment specifications. More descriptions, precautions and mitigation measures can be found in the IEC 62305 series.
- 17. Publication No Cu0141 Issue Date: August 2015 Page 14 ADD-ON PROTECTION SPD APPLICATION AT SERVICE ENTRANCE OR POINT OF USE Experience has indicated that built-in surge protection, as designed by equipment manufacturers, is not always sufficient. Efforts have been applied to correct this situation through the development of IEC standards on equipment immunity. However, there is much equipment in use that was built before these immunity standards emerged. Furthermore, many users wish to ensure more reliability and therefore routinely prescribe add-on surge protection. This desire leads to the need for selecting the most effective point of application of SPDs – service entrance, point of use, or both. The term service entrance is used in this Guide as a general term. The actual installation described by this term might have different implementations, depending upon local codes and practices. The interpretation of the term also depends on the type of service connection (overhead up to the building, underground for some distance before the building) and voltage rating of the service connection. The following discussion is presented for the case of power being supplied to the customer at a low voltage, and the service entrance understood as the point of common coupling. In such an arrangement, the user has discretionary control on what SPD may be installed at the service entrance. The selection of a two-stage protection is also at the user's discretion. Large commercial installations are generally implemented under the guidance or supervision of a competent facility engineer. At the lower end of the power scale, residential installations are generally implemented with little engineering consideration applied to the choice made by the resident on providing surge protection. An exception to this situation is emerging, with some utilities offering a packaged whole house protection engineered with coordinated service entrance and point-of-use (plug-in) SPDs. This engineered two-stage scheme takes into consideration the issues of cascade coordination, a topic that is discussed in the next subsection. Providing well-coordinated SPDs at the service entrance diverts impinging surge currents to earth at that point, rather than allowing them to flow into the installation wiring where they can induce interfering voltages in the circuits and equipment. The selection of SPDs installed at the service entrance, upstream from the customer's overcurrent protection, implies that the SPDs incorporate a suitable SPD disconnector rated for the available fault current that might develop in case of SPD failure. CASCADE COORDINATION GENERAL OBJECTIVE OF COORDINATION A coordination study of SPDs and equipment is needed whenever more than one SPD is used to protect equipment. This study should be executed after it has been verified that the protection level of the SPDs and their location are suitable for the equipment to be protected. The general objective of the coordination study is to reduce the overvoltages by means of SPDs to the withstand capability of the equipment to be protected and to ensure that the surge current rating capability of the individual SPDs will not be exceeded. This condition can be expressed as follows: Given two SPDs connected in parallel, SPD1 and SPD2 and separated by a decoupling impedance (Figure 6), the energy coordination is achieved if the portion of energy dissipated through SPD2 is lower than or equal to the maximum energy-withstand of SPD2 for each surge-current level and waveform to be considered.
- 18. Publication No Cu0141 Issue Date: August 2015 Page 15 Figure 6 Coordination between two SPDs can be demonstrated by five different approaches, described below. Whatever the approach, it is necessary that the end-user have control over the selection of the SPDs, preferably both of them. If one is already given and beyond the control of the end-user, the coordination of the second candidate SPD can be assessed and more suitable candidates identified if necessary. Most of the time, coordination is, or might seem to be a complex situation. Some of the following five procedures might then appear to the end-user as difficult to apply, due to lack of knowledge concerning some data, as for example accurate SPD characteristics. Nevertheless, coordination is necessary to ensure technically effective and cost-effective use of resources. ASSESSMENT OF COORDINATION The five basic different methods of assessing the coordination of two SPDs are: 1. Application of preferred SPD combinations. This is the most convenient variant for the users because the burden of demonstration is incumbent upon the SPD manufacturer. However, this approach implies either that a sole source manufacturer be selected, or that standard methods be developed to allow determination that candidate devices from different sources are in fact equivalent. 2. Coordination computation. Computer simulation enables complex systems to be examined and parametric evaluation performed over a wide range. 3. Application of the Let-Through Energy concept. In this concept, some decoupling impedance is postulated in the upstream SPD and a computation is performed for converting the SPD characteristics on the basis of an equivalent combination wave generator. A comparison can then be made with the energy-withstand capability of the downstream SPD (Hasse et al., 1994] 4. Coordination test. A full-scale test is performed with candidate SPDs, for a postulated decoupling impedance and a range of surge currents such that blind spots, if any, are revealed. Candidate devices may include voltage-limiting SPDs, voltage-switching SPDs, or combination type SPD [voltage- switching SPD + voltage-limiting SPD]. 5. Simplified rules with margin. Simplified tables including margins for the coordination of some typical SPDs may be used when no other data is known from the SPD manufacturers. The values given in these tables include sufficient margins to cover discrepancies between manufacturers and manufacturing tolerances.
- 19. Publication No Cu0141 Issue Date: August 2015 Page 16 SYSTEM INTERACTION PROTECTION As discussed in Cu0138, interactions at equipment ports involving different systems can produce damaging or upsetting shifts in reference potentials during surge events. Anecdotal information indicates that this mechanism is probably the most frequent cause of electronic consumer equipment failure (TV receivers and VCRs with cable signal input, computers with wired modem connection, smart telephones, etc.) Various mitigation schemes have been proposed to remedy upsetting or damaging potential differences. Increasing the withstand capability of the PC system by the manufacturers is unlikely to occur, given the market economics. Moreover, it actually might not be practical for some of the voltages that can appear in actual situations. The most effective course would be a fibre optic decoupling, inserted in the communications link, but the expense and involvement required would be objectionable for the typical home office application. For industrial applications where the consequences can be severe, this approach is often selected and has been found to be both successful and cost-effective NON-METALLIC LINKS Consumer options are limited to either replacement of existing conducted communication and control links with their fibre optic counterparts, or insertion of commercially available isolation devices as needed. The first option, while appropriate for new installations, might not be cost-effective for an existing installation. Devices manufactured specifically for various port types can be applied for existing installations. For example, RS-232 communications are often used for connecting programmable logic controllers and other devices in utility and industrial facilities. These links are easily isolated using an optically coupled adapter that connects directly to the RS-232 port. These devices must have a suitable voltage isolation rating (e.g. kilovolts) for the application. Some of these devices only have isolation capability measured in hundreds of volts which might not be sufficient. Similar devices exist for other ports. EQUALIZING REFERENCE POTENTIALS Theoretically, a simple way to eliminate the shifts in reference potentials during surge events would be to bond the two reference conductors of the two systems. However, such a simple bonding might not be acceptable from the point of view of the system operators, the local or national codes, and product listing agencies. Therefore, this Guide cannot provide recommendations applicable to all situations in all countries, but only describe general principles leading to a solution. When bonding the two references directly is not permitted, it might be permissible to provide an SPD of appropriate characteristics that will in effect bond the two references – although with some potential difference still left – but only during a surge event. This legitimately circumvents the prohibition of a permanent bonding under steady-state conditions. The alternative, i.e. no bond at all, is that unintended bonding is likely to occur – with damaging results – by sparkover of the clearances separating the conductors of the two systems. For instance, tests have shown that the isolation between the cable input and the tuner of a typical TV receiver will flashover at about 2.5 kV [Martzloff, 2000].