Editing Surge protector (section)

== Important specifications ==
[[File:Single-outlet surge protector.jpg|thumb|upright|Single-outlet surge protector, with visible connection and protection lights]]
These are some of the most prominently featured specifications which define a surge protector for AC mains, as well as for some data communications protection applications.

=== Clamping voltage ===
Also known as the '''let-through voltage''', this specifies what spike voltage will cause the protective components inside a surge protector to short or clamp.<ref name="Grouper_C">{{cite web|url=http://grouper.ieee.org/groups/spd/html/terms_c.html|title=Terms C|website=grouper.IEEE.org|access-date=18 January 2018|archive-url=https://web.archive.org/web/20160303172757/http://grouper.ieee.org/groups/spd/html/terms_c.html|archive-date=3 March 2016|url-status=dead|df=dmy-all}}</ref> A lower clamping voltage indicates better protection, but can sometimes result in a shorter life expectancy for the overall protective system. The lowest three levels of protection defined in the [[UL rating]] are 330 V, 400 V and 500&nbsp;V. The standard let-through voltage for 120 V AC devices is 330 volts.

[[Underwriters Laboratories]] (UL),<ref>{{cite web|url=http://www.ul.com/global/eng/pages/corporate/aboutul|title=About UL |date=18 July 2014|website=UL.com|access-date=18 January 2018}}</ref> a global independent safety science company, defines how a protector may be used safely. UL 1449 became compliance mandatory in jurisdictions that adopted the NEC with the 3rd edition in September 2009 to increase safety compared to products conforming to the 2nd edition. A measured limiting voltage test, using six times higher current (and energy), defines a voltage protection rating (VPR). For a specific protector, this voltage may be higher compared to a Suppressed Voltage Ratings (SVR) in previous editions that measured let-through voltage with less current. Due to non-linear characteristics of protectors, let-through voltages defined by 2nd edition and 3rd edition testing are not comparable.<ref name="SiemensUL1449">{{Cite web |url=http://w3.usa.siemens.com/us/internet-dms/btlv/PowerDistributionComm/PowerDistribution/docs_EABU%20docs/UL1449_3rd_Edition%20Revised.pdf |title=UL 1449, 3rd ed.: SPD/TVSS Changes Effective September 29, 2009}}</ref>

A protector may be larger to obtain a same let-through voltage during 3rd edition testing. Therefore, a 3rd edition or later protector should provide superior safety with increased life expectancy.

A protector with a higher let-through voltage, e.g. 400 V vs 330 V, will pass a higher voltage to the connected device. The design of the connected device determines whether this pass-through spike will cause damage. Motors and mechanical devices are usually not affected. Some (especially older) electronic parts, like chargers, LED or CFL bulbs and computerized appliances are sensitive and can be compromised and have their life reduced.

=== Joule rating ===
[[File:SpikeStopperHomeWiki.jpg|alt=|thumb|A surge protection device mounted on a residential circuit breaker panel]]
[[File:Varistorfail full.jpg|thumb|A varistor inside a consumer-grade surge protector has failed after a lightning strike]]
The Joule rating number defines how much energy a MOV-based surge protector can theoretically absorb in a single event, without failure. Better protectors exceed ratings of 1,000 joules and 40,000 amperes. Since the actual duration of a spike is only about 10 microseconds{{citation needed|date=June 2019}}, the actual dissipated energy is low. Any more than that and the MOV will fuse, or sometimes short and melt, hopefully blowing a fuse, disconnecting itself from the circuit.

The MOV (or other shorting device) requires resistance in the supply line in order to limit the voltage. For large, low resistance power lines a higher joule rated MOV is required. Inside a house, with smaller wires that have more resistance, a smaller MOV is acceptable.

Every time an MOV shorts, its internal structure is changed and its threshold voltage reduced slightly. After many spikes the threshold voltage can reduce enough to be near the line voltage, i.e. 120 vac or 240 vac. At this point, the MOV will partially conduct and heat up and eventually fail, sometimes in a dramatic meltdown or even a fire. Most modern surge protectors have circuit breakers and temperature fuses to prevent serious consequences. Many also have an LED light to indicate if the MOVs are still functioning.
 
The joule rating is commonly quoted for comparing MOV-based surge protectors. An average surge (spike) is of short duration, lasting for nanoseconds to microseconds, and experimentally modeled surge energy can be less than 100 joules.<ref>{{cite web|url=http://www.eeel.nist.gov/817/pubs/spd-anthology/files/No%20joules.pdf|title=No Joules for Surges: Relevant and Realistic Assessment of Surge Stress Threats|website=NIST.gov|access-date=18 January 2018|archive-url=https://web.archive.org/web/20130225210416/http://www.eeel.nist.gov/817/pubs/spd-anthology/files/No%20joules.pdf|archive-date=2013-02-25|url-status=dead}}</ref> Well-designed surge protectors consider the resistance of the lines that supply the power, the chance of lightning or other seriously energetic spike, and specify the MOVs accordingly. A little battery charger might include a MOV of only 1 watt, whereas a surge strip will have a 20 watt MOV or several of them in parallel. A house protector will have a large block-type MOV.

Some manufacturers commonly design higher joule-rated surge protectors by connecting multiple MOVs in parallel and this can produce a misleading rating. Since individual MOVs have slightly different voltage thresholds and non-linear responses when exposed to the same voltage curve, any given MOV might be more sensitive than others. This can cause one MOV in a group to conduct more (a phenomenon called [[current hogging]]), leading to possible overuse and eventual premature failure of that component. However the other MOVs in the group do help a little as they start to conduct as the voltage continues to rise as it does since a MOV does not have a sharp threshold. It may start to short at 270 volts but not reach full short until 450 or more volts. A second MOV might start at 290 volts and another at 320 volts so they all can help clamp the voltage, and at full current there is a series ballast effect that improves current sharing, but stating the actual joule rating as the sum of all the individual MOVs does not accurately reflect the total clamping ability. The first MOV may bear more of the burden and fail earlier.

One MOV manufacturer recommends using fewer but bigger MOVs (e.g. 60&nbsp;mm vs 40&nbsp;mm diameter) if they can fit in the device. It is further recommended that multiple smaller MOVs be matched and derated. In some cases, it may take four 40&nbsp;mm MOVs to be equivalent to one 60&nbsp;mm MOV.<ref name="Walaszczyk0"/>

A further problem is that if a single inline fuse is placed in series with a group of paralleled MOVs as a disconnect safety feature, it will open and disconnect all remaining working MOVs.

The ''effective'' surge energy absorption capacity of the entire system is dependent on the MOV matching so derating by 20% or more is usually required. This limitation can be managed by using carefully ''matched sets'' of MOVs, matched according to manufacturer's specification.<ref name="LittelfuseEC638">{{cite web |last= |title=EC638 – Littelfuse Varistor Design Examples |url=http://www.littelfuse.com/data/en/Application_Notes/EC638.pdf |access-date=2011-03-29 |publisher=Littelfuse, Incorporated}} See pp. 7–8, "Parallel Operation of Varistors".</ref><ref name="Walaszczyk0">{{cite web |url=http://www.littelfuse.com/data/en/Technical_Articles/Littelfuse_SizingMOVs_EC921.pdf |title=Walaszczyk, et al. 2001 "Does Size Really Matter? An Exploration of ... Paralleling Multiple Lower Energy Movs". |website=Littelfuse.com |access-date=18 January 2018}}</ref>

According to industry testing standards, based on [[IEEE]] and [[ANSI]] assumptions, power line surges inside a building can be up to 6,000 volts and 3,000 amperes, and deliver up to 90 joules of energy, including surges from external sources not including lightning strikes.

The common assumptions regarding lightning specifically, based ANSI/IEEE C62.41 and UL 1449 (3rd ed.) at time of this writing, are that minimum lightning-based power line surges inside a building are typically 10,000 amperes or 10 kiloamperes (kA). This is based on 20&nbsp;kA striking a power line, the imparted current then traveling equally in both directions on the power line with the resulting 10&nbsp;kA traveling into the building or home. These assumptions are based on an average approximation for testing minimum standards. While 10&nbsp;kA is typically good enough for minimum protection against lightning strikes, it is possible for a lightning strike to impart up to 200&nbsp;kA to a power line with 100&nbsp;kA traveling in each direction.

Lightning and other high-energy transient voltage surges can be suppressed with pole-mounted suppressors by the utility, or with an owner-supplied whole-house surge protector. A whole-house product is more expensive than simple single-outlet surge protectors and often needs professional installation on the incoming electrical power feed; however, they prevent power line spikes from entering the house. Damage from direct lightning strikes via other paths, such as telephone lines, must be controlled separately.

=== Response time ===
[[File:Lightningarrestor.jpg|thumb|This typical low-power lightning protection circuit combines fast-acting MOVs (blue disks) with higher-capacity GDTs (small silver cylinders).]]
Surge protectors do not operate instantly; a slight delay exists, some few nanoseconds. With longer response time and depending on system impedance, the connected equipment may be exposed to some of the surge. However, surges typically are much slower and take around a few [[microsecond]]s to reach their peak voltage, and a surge protector with a [[nanosecond]] response time would kick in fast enough to suppress the most damaging portion of the spike.<ref name="Grouper_R">{{cite web |url=http://grouper.ieee.org/groups/spd/html/terms_r.html |title=Terms R |website=grouper.IEEE.org |access-date=18 January 2018 |archive-url=https://web.archive.org/web/20170409072210/http://grouper.ieee.org/groups/spd/html/terms_r.html |archive-date=9 April 2017 |url-status=dead|df=dmy-all}}</ref>

Thus response time under standard testing is not a useful measure of a surge protector's ability when comparing MOV devices. All MOVs have response times measured in nanoseconds, while test waveforms usually used to design and calibrate surge protectors are all based on modeled waveforms of surges measured in microseconds. As a result, MOV-based protectors have no trouble producing impressive response-time specs.

Slower-responding technologies (notably, GDTs) may have difficulty protecting against fast spikes. Therefore, good designs incorporating slower but otherwise useful technologies usually combine them with faster-acting components, to provide more comprehensive protection.<ref name="LittelfuseEC640"/>

[[Image:SwitchboardSurgeProtector-40kA-DinRail.jpg|thumb|right|A two-pole surge protector for installation in [[distribution board]]s]]

=== Standards ===
Some frequently listed standards include:
* [[IEC]] 61643-11 Low-voltage surge protective devices – Part 11: Surge protective devices connected to low-voltage power systems – Requirements and test methods (replaces IEC 61643-1)
* [[IEC]] 61643-21 Low voltage surge protective devices – Part 21: Surge protective devices connected to telecommunications and signalling networks – Performance requirements and testing methods
* [[IEC]] 61643-22 Low-voltage surge protective devices – Part 22: Surge protective devices connected to telecommunications and signalling networks – Selection and application principles
* [[CENELEC|EN]] {{not a typo|61643-11}}, 61643-21 and {{not a typo|61643-22}}
* [[Telcordia Technologies]] Technical Reference [https://telecom-info.njdepot.ericsson.net/site-cgi/ido/docs.cgi?ID=SEARCH&DOCUMENT=TR-NWT-001011& TR-NWT-001011]
* [[ANSI]]/[[IEEE]] C62.xx
* [[Underwriters Laboratories]] (UL) 1449
* [[AS/NZS]] 1768

Each standard defines different protector characteristics, test vectors, or operational purpose.

The 3rd Edition of UL Standard 1449 for SPDs was a major rewrite of previous editions, and was also accepted as an ANSI standard for the first time.<ref name="Siemens1449">{{cite web|last=Siemens AG|title=Next Generation Surge Protection: UL 1449 Third Edition|url=http://www.sea.siemens.com/us/internet-dms/btlv/PowerDistributionComm/PowerDistribution/docs_EABU%20docs/Next_Generation_Surge_Protection.pdf|publisher=Siemens AG|access-date=2011-03-29|archive-url=https://web.archive.org/web/20110721173031/http://www.sea.siemens.com/us/internet-dms/btlv/PowerDistributionComm/PowerDistribution/docs_EABU%20docs/Next_Generation_Surge_Protection.pdf|archive-date=2011-07-21|url-status=dead}}</ref> A subsequent revision in 2015 included the addition of low-voltage circuits for [[USB]] charging ports and associated batteries.<ref>{{cite web|title=Standard 1449 &ndash; Standard for Surge Protective Devices|url=http://ulstandards.ul.com/standard/?id=1449|publisher=UL LLC|access-date=February 18, 2016}}</ref><ref>{{cite web|url=http://incompliancemag.com/ul-publishes-new-edition-of-ul-1449/|website=In Compliance Magazine|title=UL Publishes New Edition of UL 1449|date=2 September 2014 |access-date=February 18, 2016}}</ref>

EN 62305 and ANSI/IEEE C62.xx define what spikes a protector might be expected to divert. EN 61643-11 and 61643-21 specify both the product's performance and safety requirements. In contrast, the IEC only writes standards and does not certify any particular product as meeting those standards. IEC Standards are used by members of the CB Scheme of international agreements to test and certify products for safety compliance.

None of those standards guarantee that a protector will provide proper protection in a given application. Each standard defines what a protector should do or might accomplish, based on standardized tests that may or may not correlate to conditions present in a particular real-world situation. A specialized engineering analysis may be needed to provide sufficient protection, especially in situations of high [[lightning]] risk.

In addition, the following standards are not standards for standalone surge protectors, but are instead meant for testing surge immunity in electrical and electronic equipment as a whole. Thus, they're frequently used in the design and test of surge protection circuitry.

* [[IEC]] [[IEC 61000-4-2|61000-4-2]] Electrostatic discharge immunity test
* [[IEC]] [[IEC 61000-4-4|61000-4-4]] Electrical fast transient/burst immunity test
* [[IEC]] [[IEC 61000-4-5|61000-4-5]] Surge immunity test