電弧爆炸

Electric ArcExplosions

電弧爆炸

Vytenis Babrauskas

Abstract

摘要

When an electric arc is created, apressure event occurs. There can be two aspects to this: the shock and soundwaves propagated from the expanding arc channel, and the bulk pressurization ofthe enclosure, if arcing is taking place within a closed volume. The presentpaper is the first systematic review of the research on both these pressurephenomena. Quantitative studies on electrical arc explosion pressures date backto the 1920s, although arc pressures generated by lightning, which is a type ofelectric arc discharge, have been studied since the 1700s, but understanding ofthe phenomena is still not complete or exhaustive. Experimental data arecompared to theoretical predictions. It is shown that in an enclosed volumesome extremely high pressures can be generated, if the arc current issufficient. Such pressures can destroy buildings and mechanical equipment andcause injuries or death to nearby individuals. Even without enclosures, theshock waves produced from high energy arcs can cause injuries, although arcflash injury may be of greater concern. Injury potential generally requiresthat high currents be available, and serious damages or injuries are notassociated with low energy arcing occurrences.

電弧產生時會發生壓力變化。這種壓力變化主要有兩方面原因：一是電弧通道膨脹，產生衝擊波和聲波；二是在封閉空間內拉弧，會造成空間壓力增大。本文首次對這兩種產生壓力變化的現象進行了系統地綜述。閃電是電弧放電的一種類型，也會產生壓力變化，此方面的研究可以追溯至18世紀。關於電弧爆炸壓力的定量研究則可以追溯到20世紀20年代，但是對這種現象的認識仍不夠深入或不全面。比較實驗數據與理論推測的結果表明，如果電弧電流足夠大，會在封閉空間內產生極高的壓力。這種壓力可能造成建築和機械設備的損壞，甚至造成周圍人員的傷亡。雖然電弧閃燃造成的傷亡引起的關注更多，但高能電弧產生的衝擊波也可能造成人身傷害，即使在沒有空間限制的情況下這種傷害也可能發生。要造成此種傷害，往往需要較大的電流，低能電弧通常不會造成嚴重破壞或傷害。

Keywords：arc pressure, electric arcs, electrical accidents,explosions, shock waves, switchgear

關鍵詞：電弧壓力；電弧；電氣事故；爆炸；衝擊波；開關設備

Introduction

Electric arcing in circuits with sizablemaximum short-circuit current capacity can be a highly energetic effect. Infact, buildings have collapsed due to arc pressure, since in an enclosed spacesome surprisingly large pressures can be built up. 『Arc flash』 is the thermalradiation component associated with energetic electric arcs, and it hasreceived a great deal of study in recent years since thermal radiation has beena cause of serious burn injuries to electricians1 . Consequently, computationalmethods and research have focused on the design of appropriate protectiveclothing. For arc pressures, however, no comparable activity has taken place.In fact, the literature is sparse and not systematic on this topic. It is thepurpose of this paper to provide the first-ever review of electric arcexplosions. The emphasis is placed on pressures developed and on thecalculational methods available for these, along with experimental data thathave been published. Some of the results are strikingly high. For instance, inone test explosion overpressures of 83 atm were obtained. The magnitude of thiscan best be appreciated by considering that a fuel-air deflagration willtypically attain only around 7 – 8 atm, barring pressure-piling effects orother turbulence enhancements. During normal operation of a circuit breaker,arc pressures of roughly 3 atm magnitude can be expected2 , but these devicesare designed to sustain the pressures generated by the normal arcing associatedwith circuit opening.短路最高電流值較大的電路中產生的電弧，可能呈現高能特徵。在封閉空間中產生電弧，導致壓力急劇增大，甚至可能造成建築倒塌。「電弧閃燃」是與高能電弧有關的熱輻射。由於熱輻射是造成電工嚴重傷害的原因之一，近年來進行了大量的相關研究工作[1]。因此，關於此方面研究及計算方法，主要集中在防護服的設計上。然而，對於電弧壓力的研究，並沒有引起相當的重視。實際上，有關此方面的文獻非常少，而且缺乏系統性。本文首次對電弧爆炸進行了綜述研究。文中通過已經公開的實驗研究，重點介紹壓力的產生及相關計算方法。其中，有些結果顯示電弧爆炸產生的壓力還是相當高的。例如，在一項試驗中，爆炸壓力達到83個大氣壓。然而，如果沒有壓力積聚或湍流增強現象參與，可燃物與空氣混合物爆燃只能達到約7-8個大氣壓，據此可以更好地了解電弧爆炸的量級。在斷路器正常運行期間，電弧產生的壓力預計可以達到3個大氣壓左右[2]。按照設計要求，設備是可以承受電路正常斷開時產生的電弧壓力的。With regards to the energy supplying thearc, arcs can be of three types: (1) discharge of a fixed amount of storedenergy (e.g., a capacitor; a current transformer); (2) DC power sources; or (3)AC power sources. Lightning strikes are the most important form of storedenergy discharge, since capacitive discharge tends to be confined tospecialized situations and is uncommon as a source of industrial accidents.Since most of the power transmission and distribution networks are AC, the bulkof the research available has focused on AC arcs. But heavy-power DC systemsalso exist and are important in certain industries (e.g., electric trainpropulsion). DC arc explosions are fundamentally different since there is no『zero-crossing』 in DC. In AC circuits, an arc will extinguish at 2× the power frequency(e.g., at 100 or 120 Hz), although it may reignite very shortly afterwards. InDC circuits, this extinguishing characteristic does not exist, and arcs willgenerally extinguish only due to external circuit interruption or due toexcessive electrode consumption. Most aspects of arc behavior only depend onarc current and arc power, and not on the type of power supply, but where thetype of power supply does matter, this will be considered. The discharge of afixed amount of energy is termed a spark, while a sustained discharge is an arc, but againthis distinction will generally be seen not to affect the analysis of results.電弧按照能量供給形式可分為三種類型：（1）固定存儲的能量放電（例如：電容器、電流互感器）；（2）直流電源；（3）交流電源。雷電是最重要的存儲能量放電形式。而電容放電則因其在特定情況下才能發生，很少導致工業事故。由於多數輸配電網都是交流電（AC），因此目前已掌握的多數研究都集中在交流電弧上。但也存在大功率直流系統，在有些產業發揮著重要作用（例如：電動火車動力系統）。由於在直流系統沒有電流歸零，直流電弧爆炸與交流電弧爆炸存在本質不同。在交流電路中，電弧會以2倍的電源頻率（例如：100或120Hz）熄滅，但其熄滅後仍會復燃。在直流電路中，不存在這種熄滅特徵，通常只有外部電路中斷或由於電極過度損耗。多數電弧行為只取決於電弧電流和電弧功率，而不是電源類型。但是如果電源類型明確，可以將其作為考慮因素。固定能量放電稱之為電火花（spark），而持續能量放電稱之為電弧(arc)，但這種區別通常不會影響結果分析。Arc explosions are not rare in industry,and in other situations where 480 volt, or higher, voltages are utilized, butpublished case histories are scarce. Neither of the two large electricalaccident compilations3,4 mentions the subject. Lee5 published four brief casehistories, Crawford et al.6 documented seven case histories of arc explosionsinside motor terminal boxes, include one fatality, while Heberlein et al.7 describedtwo non-fatal explosions inside motor control centers. The best-known incident wasin an Atlanta high-rise building that took place on 30 June 1989. The fumblingof an electrician replacing a fuse caused a 480 VAC bus duct explosion8 and theexplosion and subsequent fire led to five fatalities. Lightning strikes canlead to arc explosions in any type of premises. In 1773, Lind demonstrated thatif a conductor from a lightning arrester is run down through a house, but witha small gap in this conductor, this can form a spark gap and a strike to thearrester can result in an arc explosion capable of destroying the house9 .Individuals have been bodily knocked over when in proximity both to electricalfault arcs and lightning strikes, although interestingly often there have been negligibleinjuries to the individual knocked over10. But in cases where roofs collapse,the outcome may be traumatic if persons are present underneath.在工業場所以及電壓超過480V的用電場所，電弧爆炸並不罕見，但是目前公開的事故記錄非常少[3,4]。兩大電氣事故案例彙編都沒有涉及此類事故。Lee[5]介紹了四個簡短的事故案例。Crawford[6]等記錄了七起電動機接線盒內發生的電弧爆炸事故，其中包括一起亡人事故，Heberlein[7]等介紹了電動機控制中心內發生的兩次非傷亡性爆炸。最著名的事故是1989年6月30日在亞特蘭大一幢高層建築中發生的。電工更換保險絲動作失誤，導致480V AC母線溝槽發生爆炸並引發火災，造成五人死亡[8]。在各類場所中，雷擊都可能引起電弧爆炸。1773年，Lind研究證明，對於從房屋中穿過的避雷器導體，由於其內部存在間隙，形成了電弧間隙，當雷擊中避雷器時，電弧爆炸就會發生，造成房屋損毀[9]。當人靠近故障電弧和雷擊時，可能被擊倒，但是有趣的是，擊倒造成的傷害較小，往往被忽略[10]。如果是屋頂倒塌，人在下面，後果則不堪設想。Eardrum rupture can be expected atexplosion overpressures of 19 kPa (10% probability) or 45 kPa (50%probability), while death due to lung damage is 120 kPa (10% probability) or141 kPa (50% probability). The above values come from an extensive statisticalstudy by Eisenberg et al.11; older data are somewhat different, but notgreatly. In any case, they indicate that it does not take large over pressuresfor injury or death to result from explosion pressures.當爆炸壓力達到19 kPa（概率為10％）或45 kPa（概率為50％）時，耳膜可能出現破裂，當爆炸壓力達到120 kPa（概率為10％）或141kPa（概率為50％）時，將會因肺部損傷而導致死亡。以上結果來自Eisenberg[11]等人的大量數據統計研究；舊數據有些不同，但差別不大。無論如何，都表明電弧爆炸產生的超壓，無法造成人員傷亡。Only arc explosions in gases will be reviewedhere, even though arc explosions in electrical insulating liquids can be ofimportance and arc explosions underwater are of some specialized interest.Apart from true arc explosions, conventional fuel-air explosions can also arisedue to electrical causes. Perhaps the most common type of explosion associatedwith electricity is where an electric spark ignites a flammable gas mixture ordust cloud. These explosions can be severe and destructive, but are not coveredhere, because the electric power does not provide the energy for the explosion,instead, it is the chemical oxidation reaction which serves as the energysource. Also excluded are explosions in manholes, underground ducts, andsimilar installations which may entail both fuel-air explosion and arcexplosion components. Transformer explosions12 also involve some arc explosionsaspects, but they are typically complicated and will likewise not be coveredhere.儘管在絕緣液體中發生電弧爆炸較為重要，在水下發生電弧爆炸也有特殊意義，但這裡僅介紹氣體中發生的電弧爆炸。除了真正的電弧爆炸外，電氣故障也可能引起可燃氣體與空氣的混合氣體爆炸。或許，與電氣關聯最緊密的爆炸類型是電火花引燃易燃氣體混合物或粉塵。這些爆炸可能會很嚴重，且具有破壞性，但本文不涉及這些爆炸，因為電做功不為爆炸提供能量，而是化學氧化反應作為能量來源。本文也不涉及檢查井、地下管道和類似設施中發生的爆炸，這些爆炸可能同時引起燃油-空氣爆炸和電弧爆炸。變壓器爆炸[12]也包含電弧爆炸，但此類爆炸通常較為複雜，同樣此處不再涉及。An arc explosion arises due a very rapidheating of air or other medium. In the process, electrical energy is convertedinto other forms of energy: dissociation (breaking up of molecules, e.g.,separating O2 into 2O), ionization, and heating of the gas, including itscompression; thermal radiation; and conduction losses into adjacent solids suchas electrodes. In addition, some electrode metal is vaporized and thiscontributes to the total volume which is being explosively heated, yet, therole of chemical reactions has only recently been explored. Thunder is anacoustic manifestation of an electric arc explosion and has been studied forcenturies (the currents involved in a lightning discharge are typically 20 –200 kA, which is in the same range as for serious industrial electric arcaccidents). Yet only in the last decade has there been some understandingachieved of the relative importance of the mechanisms involved, primarily dueto Graneau and coworkers13. They demonstrated that thunder could not beadequately explained by thermal or electrodynamic effects, and this conclusionwould pertain to arc explosions in general. In one experiment, they used an 8µs duration capacitive discharge and showed that the energy measured from thearc was 124% of the energy stored in the capacitor. The 24% gain was not ameasurement error, but represented chemical energy liberated by breaking chemicalbonds. Specific pathways were found to include dissociating N2 and O2 moleculesinto atoms, then forming products such as O3 and NOX. Researchers havegenerally ignored the need to account for stored chemical energy, thuspublished heat balance calculations are likely to be subject to error, especiallyfor short-duration arcs. Graneau et al. also noted that in the lightningcommunity, high current, short-duration discharges are termed 『cold lightning.』This is not hyperbole and they found that while discharges mechanically tore apiece of paper inserted into the path of the arc, they did not ignite the papernor even char it. The paper could be ignited, however, if placed in contactwith the electrodes. They also observed that lightning hitting water did notgenerate steam.由於空氣或其他介質的急劇升溫，將導致電弧爆炸發生。此過程中，電能轉換為其他形式的能量：分解（分子分解，例如：將1個氧氣分子分解為2個氧原子）、電離、氣體加熱、氣體壓縮；熱輻射；以及熱傳導進入相鄰固體（例如電極）的損耗。另外，有些電極金屬發生氣化，這將決定爆炸加熱的總體積，但是化學反應的作用也是最近才開始研究的。雷聲是電弧爆炸的聲音表現形式，並且已經經過多個世紀的研究（在雷電放電過程中，電流通常可達20-200 kA，與嚴重的工業電弧事故在同一範圍）。但是，僅在近十年中，才由Graneau及其團隊的努力，對這一過程的機理進行了較為深入的研究[13]。他們證明，通過熱或電動力效應是無法充分解釋雷電現象的，並且得出的結論通常與電弧爆炸相關。在一次實驗中，他們使用8 μs的持續放電電容，結果測量的電弧能量是電容器中存儲能量的124％，增加的24％不是測量誤差，而是通過化學鍵的破壞釋放的化學能。發現的特殊反應路徑，包括N2和O2分子分解成原子，然後形成O3和NOX等產物。研究人員往往忽略需要考慮儲存化學能，因此造成已發表的熱平衡計算出現錯誤，特別是對於短時電弧來說。Graneau等還指出，在雷電研究領域，短時大電流放電稱為「冷雷電」。這並不誇張，他們發現這種電弧放電可將電弧路徑間的紙張撕碎，但紙張沒有被引燃，甚至沒有炭化。但如果與電極接觸，紙張將被引燃。他們還發現閃電擊中水面，也不會產生蒸汽。When an arc breakdown is initiated, energygets deposited into the arc channel at a rate much greater than can be removedfrom the area by the shock wave that is created. This causes a rapid pressurerise and, if the arc energy is sufficiently high, this will be perceived as anexplosion. For a low-energy arc, the perceived sound may simple be a 『snap,』『crackle,』 or 『pop.』 But within the scientific community there is not anagreed-upon, quantitative definition12 of the term 『explosion,』 nor are there studiesto quantify the fraction of the arc energy that gets delivered as sound energy,i.e., vibrations in the 20 Hz – 20 kHz range.當電弧擊穿發生時，能量在電弧通道中積聚速度要遠遠大於產生的衝擊波向外傳播的速度。如果電弧能量足夠高，就會導致壓力的迅速上升，被視為爆炸發生。對於低能電弧，聽到的聲音可能僅是「噼」「啪」「咔」的聲音。但在科學研究領域，對於「爆炸」的概念，並沒有一個統一而又量化的定義，也沒有對轉化為聲能的電弧能量的量化研究，如：轉為聲能的振動範圍在20Hz至20kHz。Electric arc explosions are not combustionphenomena—they are predominantly physical explosions, due to very rapid conversionof electrical energy into heat. Chemical reactions play a role, but only a supportingrole, in such explosions. Recent studies suggest that chemical reactions aremainly ones which convert air to species such as O3 and NOx. While these maycomprise oxidation, they are very different from a fuel-air explosion of anormal sort. Some of the electrode metal is also vaporized in an arc explosion,and arc temperatures are high enough so that presumably much of this metalvapor may get oxidized. However, the electrode oxidation effect isquantitatively only a small part of the heat balance in an arc explosion. Thus,first-order estimates of arc explosions treat the process solely as convertingelectrical energy to heat and ignore chemical reaction contributions.電弧爆炸並不是燃燒現象。由於是電能快速轉化為熱能，主要以物理爆炸為主。化學爆炸發揮著一定作用，但僅起配角作用。最新研究表明，化學反應主要是將空氣轉化為臭氧和氮氧化物等其他物質。雖然這些反應可能包括氧化，但他們與常見類型的可燃物與空氣混合物爆炸顯著不同。在電弧爆炸中，有些電極金屬也會蒸發，且在足夠高的電弧溫度條件下，許多金屬蒸氣可能發生氧化。然而，在電弧爆炸過程中，電極氧化作用僅佔熱量平衡的一小部分。因此，對電弧爆炸進行一階推算，僅考慮電能轉換為熱能，忽略化學反應作用。In an open environment, arc pressures willrarely be highly destructive. Theoretical modeling suggests that very highpressures may be created, but experimental studies do not bear this out. Usinga 4 MV discharge, Uman did not measure any overpressures above 10 kPa, whileDrouet and Nadeau’s work indicates that at 0.3 m from an 480 Vrms, 80 kArms arc,an overpressure of only 8.7 kPa will result. But on the other hand, if a highfault current is available and arcing takes place in a small enclosure, extremelyhigh overpressures can be reached, measured in one case at 409 atm. Arc flash,which is not considered in this paper, comprises the main injury potential inopen arcing, instead of high pressures. For arcing in small enclosures, theinitial hazard may be destruction of equipment. But serious injuries may occuras consequential damages, e.g., trauma to persons from airborne cabinet parts.In addition, arcing may continue after a cabinet failure and injuries from arcflash may then ensue.在敞開環境中，電弧壓力很難具有較強的破壞性。理論建模認為是可以產生較高壓力的，但沒有得到實驗研究支撐。Uman使用4MV放電，沒有測量到超過10kPa的超壓狀態，而Drouet和Nadeau的工作表明，在距離480Vrms，80KArms電弧0.3m處，只會產生8.7KPa的超壓。但另一方面，如果存在高故障電流，且電弧發生在小空間內，則可達到極高的超壓狀態，有次測量達到了409atm。雖然本文未涉及電弧閃燃，但電弧閃燃卻是敞開電弧發生時，造成人員傷亡的主要原因，而不是高壓。對於小空間內的電弧，最初危害可能是設備損壞。但更嚴重傷害可能作為間接傷害發生，例如，拋出的櫃體部件對人員造成的創傷。此外，在櫃體受損後，電弧可能仍在持續發生，隨後電弧閃燃可能造成後續傷害。Most accidents involve AC power. However,for arcing within small enclosures, DC arcs are more hazardous than capacitivedischarge or AC arcs. With DC arcs, arcing can continue for a long time, andduring this time the enclosure pressure will progressively rise. Withcapacitive discharge arcs, the event is intrinsically short-lived. With ACarcs, continued re-ignitions of the arc may take place, nonetheless peakoverpressures will tend to be much lower than for DC conditions. A direct,quantitative comparison between DC and AC arcs is not possible, however, sinceAC arcs do not have a unique waveform and will depend on circuitcharacteristics. While many accidents involving AC arcs take place wherevoltages are 480 VAC, or higher, destructive arcing can also occur with 120 –240 VAC sources, provided sufficient current capacity, e.g., kiloamps, isavailable.多數事故涉及的是交流電源。然而，對於小空間內的電弧，直流電弧比電容放電或交流電弧更危險。對於直流電弧，電弧可以持續較長時間，在此期間，空間內部壓力將逐漸上升。對於電容放電電弧，實際持續時間非常短暫。對於交流電弧而言，電弧持續反覆引燃，但峰值超壓往往遠低於直流。然而，在直流電弧和交流電弧之間，是無法進行直接量化比較的，因為交流電弧的波形不是一致的，這取決於電路特性。雖然許多交流電弧的事故，發生在480V AC或更高的用電場所，但如果電流容量足夠（如：千安級），120-240V AC交流電源也可能產生破壞性電弧。Very little test or modeling data exist onarcing in closed, unvented, unruptured enclosures. This would be especiallygermane to arcs where the power source is sustained, i.e., DC or AC arcs, as opposedto capacitive discharges. These conditions may not necessarily represent adirect injury potential, but the analysis is simpler and reliable data woulduseful for validating of theories. Apart from some very early studies, therehas not been significant experimental work on DC arcs. A number of industriesuse DC and studies specific to DC power supplies would be valuable.關於封閉、無開口、無裂縫的空間中發生的電弧測試或模擬數據非常少。與電容放電相比，此類數據可能與電源持續供給的電弧密切相關，如：直流電源和交流電源。這些情況不一定代表具有造成直接傷害的可能性，但簡單可靠的數據分析，有助於驗證理論。除了一些早期的研究外，對直流電弧的實驗研究還遠遠不夠。許多行業使用直流電源，專門針對直流電源的研究將很有意義。In line with other fields of engineering,in recent years it has become common to conduct research studies based on CFDmodeling. Unfortunately such studies only rarely lead to phenomenological insightsor conclusions having general applicability. In addition, unless accompanied bya careful and successful validation, the numerical results may be in seriouserror.

與其他工程領域一樣，近年來基於CFD建模開展研究，已成為一種普遍現象。但遺憾的是，相關研究很少能夠得到現象學觀點或實用性的結論。此外，如果沒有仔細和成功的驗證，數值結果可能存在嚴重錯誤。